Documentation plugins for Claude Code

The understand-anything plugin leverages AI agents to analyze, visualize, and explain codebases by orchestrating multiple skills that examine project structure, dependencies, and code semantics. It integrates grep and Node.js tools to extract codebase information and triggers understanding workflows at session start and after tool use.

OpenSpec is a specification framework for planning and implementing features with AI using a structured artifact workflow that keeps teams aligned across repos. It provides skills for exploring ideas, proposing changes, implementing tasks, verifying work, and archiving completed features with auto-generated requirements and design docs.

Archify converts plain-language descriptions or Mermaid diagrams into polished, interactive HTML diagrams with dark/light themes, trace animations, and multiple export formats (PNG, JPEG, WebP, SVG, WebM). Use it to visualize system architectures, workflows, data pipelines, state machines, and other technical or planning concepts that benefit from interactive exploration.

Context7 fetches current documentation directly from source repositories, automatically resolving library identifiers and retrieving version-specific code examples when you ask about popular frameworks and libraries. The plugin provides manual commands, auto-triggering skills, and a dedicated docs-researcher agent to keep your context fresh and accurate without relying on stale training data.

This plugin restructures Claude's code output to be ADHD-friendly by leading with actionable steps, numbering tasks, suppressing tangential information, and making progress visible. It applies 10 formatting rules via a skill hook that runs at session start to transform verbose explanations into concise, action-oriented guidance.

Scientific Agent Skills is a library of 177 procedural skills for AI agents covering biology, chemistry, medicine, physics, engineering, Earth science, data analysis, and scientific communication, following the open Agent Skills standard for use with Cursor, Claude Code, Codex, and other compatible hosts. The skills provide curated guidance for specialized scientific libraries, databases, and workflows including bioinformatics, genomics, cheminformatics, proteomics, clinical decision support, and scientific visualization.

Context7 integrates with Claude Code via a remote MCP server to fetch current documentation and code examples directly from source repositories, eliminating reliance on stale training data. The plugin provides tools to search for library identifiers and query version-specific documentation by relevance to your coding question.

This plugin recreates the deprecated Explanatory output style by injecting a SessionStart hook that instructs Claude to provide educational insights about implementation choices and codebase patterns throughout each session. The insights are formatted as highlighted blocks explaining specific decisions, conventions, and trade-offs in the code.

This plugin activates a SessionStart hook that injects instructions into every coding session, encouraging Claude to request meaningful 5-10 line code contributions at decision points rather than implementing everything automatically. It combines interactive learning with explanatory insights about implementation trade-offs, design patterns, and codebase-specific decisions.

A design productivity plugin that provides critique, design system management, UX writing, accessibility audits, research synthesis, and developer handoff workflows — all available standalone and enhanced by connecting to Figma, Linear, Asana, Notion, Slack, and other team tools via MCP. The plugin includes six slash commands and six domain skills that Claude applies automatically when relevant to design work.

This AI-powered plugin automates contract review, NDA triage, and compliance workflows for in-house legal teams by analyzing documents against customizable organizational playbooks. It integrates with Slack, Box, Egnyte, Atlassian, and DocuSign to streamline legal briefings, templated responses, and institutional knowledge management.

This plugin provides an interactive viewer for PDFs that supports annotation, form filling, signatures, and approval stamps, with all processing happening locally via an MCP server. Users can mark up contracts and documents collaboratively, then download the annotated copy without needing external APIs or services.

This Claude Code plugin provides an AI-powered product management partner that writes feature specs, manages roadmaps, synthesizes user research, analyzes competitors, and tracks product metrics through integrated connections to project management and communication tools. It includes seven core commands covering the full PM workflow: specs, roadmaps, stakeholder updates, research synthesis, competitive analysis, metrics review, and brainstorming with frameworks like RICE, MoSCoW, and Jobs-to-be-Done.

claude-obsidian is a local-first knowledge system that captures sources, creates linked Obsidian notes with provenance ledgers, and retrieves grounded answers from your vault without uploading files. It provides 15 skills for ingestion, querying, linting, and maintenance with explicit consent workflows and optional methodology modes.

This skill transforms any source material (URLs, PDFs, DOCX files, Markdown, plain text, screenshots, or pasted notes) into beautiful, single-file HTML articles using the ReActicle component protocol. It guides agents through a structured six-phase editorial process with checkpoints, theme selection, and sub-agent reviewers to produce share-ready, archive-friendly articles.

This skill enables AI agents to efficiently answer questions over local, multi-format knowledge directories (Markdown, PDF, Excel) using hierarchical index navigation and progressive retrieval that avoids loading whole files into context. It enforces a learn-before-process step for PDFs and Excel files by reading format-specific references first, then uses grep and windowed reads to iteratively narrow down answers over up to 5 rounds.

No AI Slop detects and removes 20+ patterns of artificial writing—like binary contrasts, throat-clearing openers, and faux-insight setups—while preserving the writer's personal voice and vocabulary. Users can edit drafts for sharper, more human writing, detect whether text contains AI patterns, or generate satire by requesting the most clichéd AI slop possible.

This standalone Agent Skill provides two utilities: an interactive self-assessment tool that generates personalized learning paths, and an update-threat-db skill that refreshes threat intelligence through AgentSec Triage and synchronizes coding-agent and MCP security advisories. Both skills support the Claude Code Ultimate Guide, a comprehensive resource for learning Claude Code, building reliable agents, and scaling their deployment safely.

This Agent Skill helps Codex draft, restructure, translate, and review bilingual Chinese-English architecture documentation for the Maka Agent backend. It provides guidance for technical narratives, component explanations, architecture decisions, and documentation quality checks while maintaining architectural precision.

Ars Contexta generates a personalized knowledge management system through conversational setup, creating markdown vaults with wiki links, processing pipelines, automation hooks, and domain-native documentation—all derived from how you describe your thinking and work. The plugin provides 10 skills for extracting insights, finding connections, and maintaining your vault, plus a 6-phase pipeline that mirrors Cornell note-taking with a meta-cognitive layer for continuous system evolution.

The Microsoft Learn MCP Server provides Claude direct access to official Microsoft documentation, API references, and code samples for Azure, .NET, Windows, and related technologies without hallucinations. It includes three tools for semantic search, page fetching, and code sample discovery across Microsoft's official documentation, with no authentication required.

This plugin hooks into the PreCompact event to automatically summarize file paths, root causes, and open questions, keeping context intact during code compaction. It relies on an external shell script (precompact_priorities.sh) to process and organize information before compaction occurs.

SlopMonster detects common AI writing patterns (clichéd vocabulary, sentence structures, punctuation habits, and unsubstantiated claims) in English text and scores drafts out of 5, with an editing workflow that requests revisions from alternate model families to reduce AI-like patterns. The Python checker requires no API keys or external dependencies, and integrates as a Claude Code skill via `/slopmonster` or works standalone with CLI-based editing passes using different AI writers.

The LaTeX Document Skill converts plain English descriptions into publication-ready PDF documents without requiring LaTeX knowledge, supporting 27 templates and handling tasks from resume generation to OCR-powered document conversion with 27 automation scripts. It processes handwritten notes, scanned textbooks, CSV data, and other formats through specialized pipelines that compile to polished PDFs with proper formatting, equations, charts, and diagrams.

This skill package provides two specialized capabilities: `alt-text` for generating and improving accessible alt text for data visualizations and images in pkgdown sites and Quarto documents, and `brand-yml` for creating and applying brand.yml files for consistent styling across Shiny apps (R and Python) and Quarto documents. Both skills extend Claude's abilities for accessibility and design consistency in R/Python documentation and app projects.

ClaudeForge is an automated toolkit for creating, maintaining, and syncing CLAUDE.md files across Claude Code projects with a hard 150-line cap, automatic chaining via sub-files, and embedded Karpathy behavioural guidelines. It provides lifecycle hooks (PostToolUse, Stop), three parallel audit skills for drift detection and dependency scanning, and slash commands for initialization, enhancement, and synchronization.

llmwiki transforms dormant LLM session transcripts from Claude Code, Codex CLI, Cursor, Gemini CLI, and Obsidian into a beautiful, searchable, locally-hosted knowledge base following Andrej Karpathy's LLM Wiki pattern. It generates AI-consumable exports (llms.txt, JSON-LD graph, per-page metadata) and provides semantic synthesis via optional Ollama integration.

A Claude Code plugin and agent skills toolkit that teaches Claude to author, validate, and visualize Open Knowledge Format (OKF) bundles—portable markdown knowledge with YAML frontmatter that humans and agents can both read and maintain. Includes a deterministic conformance checker, interactive graph renderer, GitHub Action support, and an MCP server for read-only knowledge access.

Architecture Studio is a local-first framework for AEC firms to build governed AI workflows with persistent studio and project memory, bundling 7 agents and 50 skills for site planning, zoning, specifications, sustainability, and materials research. The plugin initializes portable workspaces, maintains linked project records, and provides extension points for firm-specific and project-specific skills that remain user-owned outside the plugin cache.

Gauntlet extracts code structure via AST analysis and builds a knowledge graph to generate adaptive coding challenges using spaced repetition, helping developers prevent knowledge atrophy and accelerate onboarding. It supports six challenge types—multiple choice, code completion, trace, explain-why, spot-the-bug, and dependency mapping—with ML-based answer scoring and pre-commit hooks to gate commits behind knowledge checks.

Memory Palace organizes domain knowledge using spatial mnemonic techniques, building persistent "palaces" and "gardens" with multi-modal search, bidirectional linking, and automatic deduplication. It captures PR review knowledge and intercepts web requests to check local knowledge before external lookups.

Sanctum automates Git and development workflows through preflight checks, conventional commit generation, PR preparation with quality gates, and version management. It also consolidates documentation, performs scope-focused code reviews, and improves workflows through session analysis and retrospectives.

Codebase Mapper automatically creates and maintains a current map of your project structure that Claude loads at the start of each session, tracking files, flows, dependencies, and conventions across any stack. It monitors code changes throughout sessions and keeps the map in sync with incremental or full updates as needed, eliminating the need for Claude to rediscover your codebase each time.

This PostToolUse hook reminds developers to update CLAUDE.md documentation files when git status or git log reveals changes that may affect contracts, APIs, or domain structure. It integrates with the project-claude-librarian agent to prompt documentation updates before commits.

Claude Code plugin providing comprehensive guidance for implementing SEO in Nuxt 3/4 applications using all 8 official Nuxt SEO v5 modules, including robots.txt, sitemaps, Open Graph images, Schema.org structured data, and link checking. Covers installation, configuration, breaking changes, and troubleshooting across core and standalone modules with multi-language support.

Sloptrim is a local detector for AI-writing patterns that scores the first 262,144 characters of extracted prose from files (up to 512 KB plain text or 4 MB for supported archives) against 71 documented patterns, naming flagged spans for fixing before publication. It uses Python standard library only with no network or model calls, reads 20 file formats including .docx and .epub, and integrates into Claude Code via three hooks (PostToolUse, SessionStart, UserPromptSubmit).

This macOS-only Agent Skill helps Claude format Word documents by replicating typography styles (fonts, sizes, indents, line spacing, alignment, page setup, stylesheets, themes, and headers/footers) from a reference .docx template to new content through a Word → Filtered HTML → edited HTML → paste pipeline. It avoids AI-generated HTML markup errors by having Claude edit only text content while preserving inline CSS styling from Word's own exports.

A complete AI-powered patent creation and analysis system that helps inventors search prior art, draft claims, and prepare applications without legal experience through guided workflows and automated validation. It integrates 16 commands, 13 agents, and 17 skills to handle patent rule lookup, prior-art searches across 100+ million patents, document review, and diagram generation entirely within Claude Code with no external MCP server required.

Sanity Agent Toolkit provides MCP server access to Sanity projects, content, schemas, and releases, along with 8 agent skills covering content modeling, SEO/AEO, and development best practices. Includes 4 slash commands and integrates with Claude Code, Cursor, Codex, VS Code, Lovable, v0, Replit, and other MCP-compatible clients.

A Claude Code plugin specialized in ADVPL and TLPP for TOTVS Protheus development, offering code generation, ADVPL-to-TLPP migration, error diagnosis, code review with 24 best-practice rules, and automated documentation. It also provides dictionary generation (SX2, SX3, SIX, SXG, SXA, SX1, SX5, SXB, SX7) for functional consultants and supports automatic CP1252 encoding conversion for all ADVPL/TLPP file types.

cc-polymath is a skill management system that organizes 410+ production-ready development skills across 33+ domains into atomic, composable units with progressive discovery via keyword-triggered gateways. It solves context-window efficiency by loading only relevant skills on-demand rather than upfront, reducing startup overhead by 60-84% compared to monolithic knowledge bases.

prd-ce is a PRD lifecycle methodology engine that scaffolds PRD/SoT/EPIC documents to template and drives them toward completion using 4 agents and 50 specialized skills organized by stage. It governs the process with 6 hooks into the Claude Code runtime and deterministic readiness scoring.

This Claude Code plugin automatically maintains CLAUDE.md project memory files as codebases evolve, tracking file edits, deletions, and moves through hooks and updating memory in an isolated agent to preserve context window. It offers zero-config initialization, git-aware syncing, and marker-based updates that preserve manual content while keeping build commands, architecture, and conventions fresh.

This MCP server integrates Claude Code with Outline, a document management platform, enabling search, read, create, edit, and archive operations on documents and collections. It includes automatic rate limiting, threaded comments support, and backlink discovery via the Outline API.

CodeDNA extends Claude Code with an in-source AI communication protocol that embeds structured annotations (exports, used_by, rules, agent logs) directly in source files across 9 languages and 7 template engines. Four commands (/codedna:init, /codedna:check, /codedna:manifest, /codedna:impact) and four automatic hooks provide architectural context, dependency tracking, annotation enforcement, and inter-agent communication without requiring external service calls.

A Claude Code plugin that provides 18 structured workflow skills for guiding AI through design, implementation, code review, testing, and documentation phases. Includes domain-specific profiles (Go, Java, JS/TS, Kotlin, K8s, Python) with checklists, validators, and review rubrics.

This MCP server fetches the latest official library documentation from authoritative sources to ground Claude's responses in current API references. It supports popular libraries and frameworks, helping prevent outdated information and reduce hallucinations by providing real-time documentation access.

This plugin provides an MCP server for searching and fetching official Microsoft Learn and Azure documentation directly within Claude Code. It includes tools for documentation search, code sample discovery, and full-page retrieval covering Microsoft, Azure, .NET, C#, and Visual Studio resources.

This Claude Code plugin automatically generates and maintains a fractal multi-level documentation system inspired by Gödel, Escher, Bach, creating self-referential PROJECT_INDEX, FOLDER_INDEX, and file-header annotations that keep code and documentation in perfect sync. It uses post-tool hooks to detect code changes and automatically regenerate the interconnected index structure across the entire project.

This standalone Agent Skill provides a complete guide to Light Protocol on Solana, covering ZK Compression for rent-free compressed tokens and PDAs using zero-knowledge proofs, plus the Light Token Program for high-performance token transactions at 200x lower cost than SPL. It includes TypeScript SDK documentation, JSON RPC methods, and integration patterns for developers building on Solana.

This skill provides comprehensive API guidance for Lulo, a Solana lending aggregator, enabling integrations for deposits, withdrawals, balance queries, and automated yield optimization across multiple protocols like Kamino, Drift, MarginFi, and Jupiter. It covers Protected/Boosted deposits and Custom deposit strategies without requiring external API keys or third-party services.

This standalone Agent Skill provides comprehensive guidance on the Metaplex Protocol for building and managing Solana NFTs and digital assets, including Core (next-gen NFTs), Token Metadata, Bubblegum (compressed NFTs), Candy Machine, Genesis (token launches), MPL-Hybrid, Inscriptions, DAS API, and the Umi framework. It serves as the authoritative source for integrating Metaplex functionality into Solana projects.

This skill provides a comprehensive guide for building high-performance Solana programs using Pinocchio, a zero-dependency, zero-copy framework, covering account validation, CPI patterns, and optimization techniques. It includes guidance on migrating from Anchor to Pinocchio for Solana developers.

This standalone Agent Skill provides a complete guide for integrating Pyth Network's decentralized oracle, enabling real-time price feeds for DeFi applications on Solana. It covers price feed integration, confidence intervals, EMA prices, on-chain CPI, off-chain fetching, and streaming updates.

This standalone Agent Skill helps Solana developers determine when to migrate from @solana/web3.js (v1) to @solana/kit, offering detailed API mappings and guidance on handling edge cases during the transition. It provides hands-on migration assistance without requiring external dependencies or API keys.

This standalone Agent Skill provides a complete guide to the Squads Protocol, Solana's leading smart account and multisig infrastructure, covering V4 Multisig for team treasury management, Smart Account Program for account abstraction and programmable wallets, and Grid for stablecoin rails and fintech infrastructure. It enables developers to understand and implement Squads' core components for secure wallet and treasury operations on Solana.

This standalone Agent Skill provides deep exploration of Solana's architecture and protocol internals, including the SVM execution engine, account model, consensus, transactions, validator economics, and development tooling. It draws from Helius blog content, SIMDs, and Agave/Firedancer source code to help understand Solana's blockchain design.

claude-mega-brain injects a compressed OKF-formatted knowledge index into Claude Code sessions automatically, enabling accurate project-specific answers without tool calls. The plugin scans for markdown files with YAML frontmatter type declarations and injects the consolidated knowledge at session start and after tool use.

This agent skill package teaches AI coding assistants how to develop ServiceNow Fluent applications by providing access to SDK documentation and project conventions through the `now-sdk-explain` command, plus environment setup via `now-sdk-setup`. The skills enable AI to look up API types, metadata conventions, and project structure before making changes to TypeScript-based ServiceNow Fluent applications.

Autoform helps formalize mathematical papers and research notes into Lean 4 proofs by planning work from source documents, writing formalized code, and reviewing progress with integrated Lean LSP and REPL tools. The plugin provides seven skills that guide users through setup, roadmap creation, formalization, and review workflows entirely within Claude Code or Codex.

A Claude Code plugin providing interactive onboarding and Q&A skills for mastering Claude Code setup, best practices, automation, and effective workflows. Includes seven skills ranging from interactive guides to introspective development audits and a gamification system that tracks feature usage across sessions.

Lisa conducts structured interviews about features using AskUserQuestion, capturing answers to generate comprehensive product requirement documents in Markdown and JSON formats. It's designed to automate specification workflows before implementation, with support for resuming interrupted sessions and optional first-principles questioning to challenge assumptions.

Bedrock automates Obsidian vault management by turning markdown notes into a structured knowledge graph with AI-powered entity detection, bidirectional linking, and sync across 7 entity types. It ingests external sources (Confluence, Google Docs, GitHub, PDFs, and more), deduplicates content, and provides orchestrated querying through Claude Code skills.

deep-plan transforms vague feature requests into detailed, production-ready implementation plans through AI-assisted research, stakeholder interviews, and multi-LLM review. The plugin orchestrates a comprehensive workflow including research, interview, external LLM review, TDD planning, and section splitting to create isolated, parallelizable implementation units.

The Documents Toolkit plugin provides fast PDF text extraction, OCR for scanned PDFs, and structured parsing of PDFs and office files (DOCX, PPTX, XLSX, ODF, EPUB, email) into JSON and markdown. It includes uv-runnable scripts for PDF forms, table extraction, merging, splitting, and validation—all with zero local installations via uvx and inline dependencies.

This Claude Code plugin provides comprehensive Shopify development reference documentation covering Liquid templating, Online Store 2.0 themes, Hydrogen framework, app development, and Functions, while deferring live GraphQL schema validation to Shopify's official tools. It includes reference materials on theme performance, debugging, API versioning, and migration guidance for deprecated features like Scripts and Hydrogen API changes.

This standalone Agent Skill identifies and eliminates characteristic patterns of Chinese AI writing—clichés, parallel repetition, nominalization, abstract subjects, aphoristic endings, and formulaic structures—by applying 8 rewriting rules and a 5-dimensional scoring system. The skill works with Claude, GPT, Gemini, and other LLMs to make AI-generated Chinese read like natural human speech across all text types: articles, social media, official documents, academia, and product copy.

This plugin analyzes Shen Lun (申论) examination materials, provides corrective feedback on responses, and assigns structured scores while enforcing safe execution. It integrates Python3-based processing through an MCP server and a PreToolUse hook for validation.

This standalone Agent Skill manages documents in self-hosted Paperless-ngx via REST API, supporting upload with OCR, full-text search, tagging, bulk operations, and metadata management. It requires a running Paperless-ngx instance and API token, and integrates with Claude through local shell scripts using curl and jq.

Wenlan is a living knowledge base that agents build during work sessions, with learned insights automatically converted into source-cited wiki pages that persist across threads. Each new session starts with the latest accumulated knowledge from previous agent work.

lintlang audits agent configs, system prompts, and tool definitions by scanning YAML, JSON, Markdown, and Python files, offering both an on-demand audit skill and automatic repair guidance after Claude Code edits supported language-bearing files. The plugin requires Python 3.10+ with lintlang installed via pipx, runs the scanner locally without making network calls, and integrates with Cursor's Agent Plugins in addition to Claude Code.

Akii is a free SEO, AEO (Answer Engine Optimization), and GEO toolkit that audits websites, optimizes pages, generates schema markup, clusters keywords, and tracks AI visibility across ChatGPT, Claude, Gemini, Perplexity, Copilot, and Google AI Overviews. The plugin includes 15 skills, 5 autonomous agents, and 8 commands for content strategy, technical audits, and multi-language localization—with optional integration of Ahrefs, GSC, Apify, and DataForSEO MCPs for richer data.

A Claude Code plugin that scaffolds and maintains a `docs/lore/` directory to capture implicit project knowledge—business rules, API quirks, pitfalls, and decision rationale—that code and git history can't reveal. Seven interlinked skills (init, consult, capture, guard, maintain, check, ul) form a self-correcting loop: capture knowledge as you learn it, consult it before acting, guard diffs against it at commit time, and health-check the lore every ~30 days to keep it alive.

Eidos enables spec-driven development in Claude Code by treating Markdown specifications as the source of truth, with code as a downstream manifestation that stays bidirectionally in sync. It provides 73 skills organized around core workflows—spec↔code loops, persistent planning and memory, research and understanding, and steering mechanisms—to keep human intent and AI reasoning aligned throughout development sessions.

LegalRabbit DOCX enables Claude to read, create, and edit DOCX files locally with tracked changes and comments, using 2-5x fewer tokens than alternatives. The plugin works completely offline to ensure data privacy and integrates seamlessly with Claude for Legal workflows.

This plugin maintains a searchable local index of Claude documentation accessible via the /docs command and auto-discovery skills. It automatically updates the documentation on session start and provides instant access to API references, guides, and tutorials without external service dependencies.

This standalone Agent Skill generates conceptual process flow diagrams (PFDs) as editable SVGs along with the Python script that created them, triggered when a user describes a process and asks to visualize it or mentions P&IDs, control loops, or cascade control. It encodes chemical manufacturing engineering standards to help Claude produce diagrams that practicing engineers would accept, explicitly disclaiming replacement for issued-for-design P&IDs.

This is a curated directory of Agent Skills—reusable instruction sets for AI coding assistants like Claude Code, Cursor, GitHub Copilot, and Gemini CLI—including guides for skill creation, platform compatibility, and contribution standards. The repo itself is a standalone skill that teaches developers how to understand, organize, and maintain the awesome-skills resource collection.

A Claude Code skill that interactively co-builds C4 architecture diagrams (Context, Container, Component, Code levels) through mode-aware dialogue, automatically detecting whether you're designing greenfield, documenting existing code/prose, reviewing, or updating. Outputs negotiable formats (Mermaid, Structurizr DSL, PlantUML) with validation at each level before writing to local filesystem or connected MCP destinations.

BetterCallClaude is a Swiss legal intelligence plugin that integrates BGE/ATF/DTF precedent research, case strategy, legal drafting, and citation verification for all 26 Swiss cantons with built-in Anwaltsgeheimnis privacy protection. It provides 27 commands across 21 agents for legal research, litigation strategy, document analysis, and multi-language legal translation (German, French, Italian, English).

claude-english-buddy automatically corrects English grammar and typos in your prompts before Claude sees them, restoring a feedback loop that helps non-native speakers improve over time. It also translates non-English input, refines prompts with a :: prefix, and tracks your error patterns and improvement trajectory through daily and long-term analytics.

This standalone Agent Skill repo provides two Dart and Flutter SDK changelog skills that equip AI coding agents with version-specific language features, APIs, widget deprecations, and migration guides spanning Dart 1.0–3.13+ and Flutter 1.0–3.47+. The skills include fast lookup matrices for minimum SDK compatibility, macros, experimental features, Material 3 migrations, and structured runbooks for rescuing legacy codebases.

Dr. Ralph implements a structured 5-phase diagnostic workflow (Interview → Research → Differential Diagnosis → Treatment Plan → SOAP Report) using Claude Code's Stop hook system for iterative medical analysis. The plugin maintains persistent patient case files and generates research-backed diagnostic reports with configurable safety limits and completion criteria.

This standalone Agent Skill generates hand-drawn animated architecture and workflow diagrams in a sketch style with ink lines, beige paper, and flowing motion that traces the topology — exporting as self-contained single-file HTML (SVG+CSS) or seamless-loop GIF. It automatically matches a fixed design system of palette, stroke, and spacing, checks its own work by re-rendering, and supports CJK text with bundled OFL handwriting fonts.

This Agent Skill applies Richard Paul and Linda Elder's 8-element thinking framework and universal intellectual standards to analyze argumentative texts, detecting logical fallacies and cognitive biases. It produces structured reports with synthesis, element-by-element analysis, fallacy detection, bias identification, and ideological symmetry testing in a conversational, pedagogical tone.

A design director plugin that generates production-ready UI/UX for marketing websites, web apps, iOS, and Android—producing HTML/CSS directly for web targets and device-framed mockups plus handoff specs for mobile. Enforces a rigorous design method covering briefs, signature elements, anti-AI-slop principles, design systems with OKLCH tokens, motion profiles, UX copy, and a deterministic review gate aligned with WCAG 2.2.

The fuse-laravel plugin is an expert Laravel 13 development assistant offering 23 specialized skills covering architecture, ORM, APIs, authentication, testing, queues, Livewire, templates, migrations, and billing with PHP 8.3+ and SOLID principles. It provides comprehensive documentation and PHP Attributes-first patterns, activated by keywords like Laravel, Eloquent, Blade, Livewire, and common development tasks.

fuse-seo provides technical SEO, schema markup, E-E-A-T validation, AI Overviews optimization, and local SEO capabilities as a local-first Claude Code plugin with framework awareness and zero third-party API dependencies. It covers sitemaps, hreflang, internal linking, content briefs, semantic clustering, and search experience optimization through 9 agents and 21 skills.

A Claude Code plugin providing expert guidance on Tailwind CSS v4.1 with CSS-native configuration, custom directives (@theme, @utility, @variant), and the faster Oxide engine. It includes 16 specialized skills covering layout, styling, utilities, responsive design, and new v4.1 features like OKLCH colors and container queries.

Houtu Project Skills is a curated collection of 2 agent skills designed to support the full project lifecycle: docs-context automatically loads project standards and syncs documentation, while houtu-dependencies guides developers in using an enterprise Java framework's modules and APIs for scenarios like session management and distributed locking. This standalone skill package requires no external API keys or paid services and installs via npx or Claude Code's plugin marketplace.

This Agent Skill provides a two-gate scanner that detects AI texture ("slop") in prose while verifying quote accuracy against source documents using pure Python with no external dependencies. It flags banned phrases, narrative clichés, and over-correction in Gate 2 (taste), then validates every quoted or attributed line against primary sources in Gate 1 (source fidelity).

knowledge-patch is an offline plugin that routes three native skills to bundled technology patches organized in compact generated catalogs, with no external downloader required. It provides setup, activation, and status workflows through Claude Code and Codex native skills, with an optional SessionStart hook to inject gateway guidance into session context.

ARIA captures session insights, decisions, and feedback into a governed, tag-indexed knowledge base that actively shapes development via change-decision hooks, contextual retrieval, and codebase mapping. The plugin enforces a five-phase lifecycle (capture → govern → promote → apply → refresh) and integrates with 10+ work-management and communication platforms via MCP servers.

This plugin detects low-value "AI slop" comments in code diffs and steers clean comment generation by extracting comments deterministically, judging them against quality criteria, and acting on each with keep/trim/rewrite verdicts. It operates in two scopes—diff-based (changes introduced) or repo-wide—and applies edits to a review branch while preserving required documentation comments and factual content.

This Claude Code plugin enforces writing style and detects AI-generated language patterns in human-facing prose like PR descriptions, comments, and documentation through hooks that check for vague vocabulary, promotional language, and templated constructions. It provides skills for analyzing writing tropes, rewriting text, scanning for violations, and conducting multi-agent document reviews using configurable wordlists and pattern detection.

This Claude Code plugin reads infrastructure-as-code files (Terraform, CloudFormation, Kubernetes, Docker, and others) to automatically generate architecture and infrastructure diagrams using D2, and keeps them synchronized with code changes via drift detection. The plugin produces multi-level diagrams (detailed and simplified) in light and dark SVG variants, with 950+ cloud and technology icons and animated traffic flow visualization.

A general-purpose Java coding plugin with 14 skills, 2 agents for architecture guidance, and tools for applying Java 8+ coding standards. It integrates with Maven and uses shell-based analysis to understand Java project structure and enforce best practices.

wiki-wonka maintains a persistent, LLM-curated knowledge base by ingesting sources into immutable raw files, generating cross-referenced Markdown pages in a wiki directory, and performing consistency checks via linting operations. The agent handles all summarization, synthesis, and bookkeeping while enforcing a schema-driven structure and language configuration across ingest, query, and lint workflows.

Renders GitHub Flavored Markdown features in Claude Code replies, including alerts, task lists, strikethrough, and Mermaid diagrams as terminal Unicode art or desktop SVG. A SessionStart hook optionally adds GFM writing guidelines to Claude's context so it naturally generates these formats.

This plugin provides hook examples and reference implementation demonstrating how to build Claude Code plugins with PreToolUse, SessionStart, and UserPromptSubmit lifecycle events. It includes four hooks with shell-based validation and utility scripts for timestamp addition and context loading.

Refine is a heavyweight multi-agent workflow plugin that cross-examines repository issues across multiple perspectives to reach consensus, then applies only the highest-ROI fixes to code and documentation. It uses a PreToolUse hook to intercept and guide tool execution, with shell capabilities for executing fixes directly.

trigger-tree measures which documentation your Claude Code instance actually discovers and whether injected instructions influence behavior, running locally with zero telemetry or token cost. It provides a live dashboard with heat maps, health diagnostics, and a CI gate that fails PRs on documentation discoverability regressions.

Haytham transforms raw startup ideas into implementation-ready specifications through a four-phase pipeline: market research and GO/NO-GO validation, MVP scoping with capability models, build-vs-buy analysis and architecture design, and OpenSpec generation ready for coding agents. It runs entirely within Claude Code using only bash and Python, with no external APIs or keys required.

The Story Driven Development (SDD) plugin synchronizes documentation with code by tracking changes through git diffs and status markers, automatically keeping stories and implementation aligned. It provides commands and hooks to integrate story-driven workflows directly into Claude Code's development environment.

Vault PKM integrates Claude Code with Obsidian, providing 20 MCP tools for structured knowledge management including semantic search, graph traversal, and enforced metadata templates. It enables bidirectional knowledge flow across conversations and projects through a unified vault with activity logging, link suggestions, and specialized agents for knowledge capture.

junior-flow helps junior developers create technical story markdown from story trackers or feature descriptions by automating the initial story-flow workflow step. It integrates with the story-flow plugin and the code-review plugin to guide developers through planning, implementation, and code review cycles.

The readable plugin injects a SessionStart hook that makes Claude format replies as styled HTML cards, with automatic tier selection (MCP Apps card server, inline widget kit, or BiDi-safe text) based on available host tools—no configuration needed. Each card loads only the CSS it uses, supports Persian/RTL with per-block direction resolution, and can include charts, tables, figures, screenshots, and looping animations that stay embedded in a single offline file.

attune-help reads and explores `.help/` documentation templates with progressive depth levels (concept → task → reference) and audience adaptation, without requiring any AI API keys. It pairs with attune-ai's authoring module or stands alone as a lightweight read-only help runtime with 4 skills and 6 MCP tools for navigating bundled and project-local documentation.

Elephant is a persistent memory system for Claude Code that retains session context, project decisions, and work history across sessions via `ELEPHANT.md` (team-shared) and global memory files. It also automates changelog generation with semantic versioning detection and README maintenance from git history and repo context.

sitemd is an MCP server plugin that turns markdown files into production-ready static websites with built-in navigation, search, themes, forms, analytics, and SEO — all configured through a local binary and settings files. It provides Claude Code with MCP tools for page management, content generation, site status, settings, and deployment to Cloudflare or any static host.

A standalone Agent Skill that provides Claude with comprehensive reference documentation for building Mintlify documentation sites, covering components, configuration, navigation, and API documentation setup. The skill intelligently routes to detailed reference files only when needed, including syntax for callouts, cards, steps, tabs, accordions, code groups, and full docs.json schema reference.

This Claude Code plugin provides comprehensive tooling for managing SPECLAN projects, helping developers organize, link, and maintain project specifications as structured markdown files. It includes 13 commands, 7 agents, and 10 skills to streamline specification authoring and navigation workflows.

This plugin assists in authoring SPECLAN specifications interactively with Claude Code, helping users craft new Goals, Features, Requirements, and Change Requests instead of editing markdown manually. It requires the speclan plugin to be installed and reuses its format, ID generation, and query capabilities.

The worklog plugin automatically captures Claude Code sessions and summarizes them into daily markdown worklogs via hooks, with no cost or LLM calls during capture. Users can view, search, and manage their work history through commands like `/worklog:show`, `/worklog:search`, and `/worklog:archive`, with optional isolated mode using a separate API key for security.

Claude of Alexandria is a biblical study and exegetical analysis plugin that retrieves linguistic data (morphology, discourse features, vocabulary, textual variants) via an MCP server and applies rigorous analytical skills using TDD methodology to prevent interpretive errors. It offers context-aware skills, a deep-study workflow with structured stages, and optional team-based discussion, with built-in guardrails to ensure Sonnet or Opus execution and prevent Haiku usage.

Dossier is an evidence-first project documentation system that produces an audit-ready 23-file package with every material claim traced to a source and verification state, enforced through three independent verification passes and a binary release gate. It ships a post-merge GitHub Actions job that keeps documentation current by opening automated documentation PRs, with built-in security isolation between privileged and unprivileged operations.

This standalone Agent Skill audits whether an app's legal documents (Privacy Policy, Terms of Service, Cookie Policy, DPA) actually reflect what the app does, checking for critical, important, and minor compliance gaps. It supports GDPR, CCPA/CPRA, LGPD, PIPL, UK GDPR, and Australian Privacy Act jurisdictions without requiring external APIs or tools.

This plugin blocks Claude's built-in Artifact tool and instead writes decision briefs as self-contained, version-controlled HTML files stored locally in your repository. The skill enforces structured writing standards—motivation first, clear definitions, inline SVG diagrams with dark-mode support—and includes automated checks for forward references and evidence marking.

This TYPO3 documentation skill guides creation and maintenance of extension documentation in reStructuredText format, with validation and Docker-based local rendering support. It provides extraction tools for analyzing code and configs, gap analysis workflows, TYPO3-specific RST directives, and automated deployment guidance through TYPO3 Intercept.

This plugin is deprecated; Claude Code v2.1.277+ natively reads AGENTS.md files through the built-in agents-md mod. It previously enabled automatic loading of AGENTS.md project instructions at session start and after file reads, with hierarchical directory support for monorepos.

bengo-toolkit is a Claude Code plugin for Japanese law firms that automates court document template filling from PDFs, generates inheritance family trees (相続関係説明図) from registry documents, proofreads legal documents, and analyzes lawsuit filings with deterministic calculation logic for statutory inheritance shares. It works best in Claude Code with calculation and legal research features also available in Claude Cowork, integrating e-Gov legislation API for statute lookup and requiring no external API keys.

gs-superadmin is a Claude Code plugin that creates a persistent Gainsight admin workspace, indexing tenant assets like rules, journeys, and scorecards into a local markdown knowledge base for citation. It provides 8 skills for drift detection, naming audits, change planning, impact analysis, and mutation-gated safety controls via the @gainsight/gs-admin-cli package.

This is a comprehensive knowledge base and agent skill for understanding, building, and auditing Claude Code mods—TypeScript add-ons that extend the Claude Code editor with panels, commands, hooks, and safety guards. It provides 100+ linked notes, a scanner for static mod analysis, step-by-step recipes, a graded catalog of 40+ mods, and the mods-secretary agent for Claude Code integration.

The researcher plugin creates distinctive, print-ready HTML and Markdown one-pagers with five curated aesthetic presets (Terminal, Editorial, Corporate, Industrial, Fresh) designed to avoid generic AI design patterns. It converts HTML documents to PDF using Chromium-based browsers and provides shell commands for batch PDF generation.

ArcKit is an Enterprise Architecture Governance Harness providing 76 slash commands across strategy, architecture, delivery, and assurance within Claude Code, leveraging 31 agents and 6 MCP servers for autonomous research and compliance enforcement. The plugin integrates with AWS, Azure, GCP, and government data sources to guide architecture decisions and maintain organizational standards through automated hooks and interactive guidance.

cv-forge converts a JSON Resume file into an ATS-friendly PDF using browser-based rendering with a choice of templates, requiring no Node, API keys, or network calls. The plugin stores your résumé as a single versioned JSON file and uses a pure shell script to splice it into HTML templates that you can print to PDF directly from your browser.

docpin grounds Claude in version-matched documentation by detecting installed library versions from your project's lockfiles and fetching docs pinned to those exact versions across npm, PyPI, crates.io, and Go. It automatically triggers before Claude writes code importing third-party libraries and provides explicit commands to pull docs for specific libraries or topics.

This skill actively builds and maintains a CONTEXT.md ubiquitous-language glossary by challenging domain terms, disambiguating overloaded words, and stress-testing relationships to ensure consistent naming across a codebase. It uses three hooks to detect repos missing a glossary and makes a one-time-per-repo offer to create one, compounding savings in token spend and code navigability over time.

fig is a Claude Code plugin that automates the non-drawing half of Figma work—preparing design files, maintaining consistency, and generating handoff documentation through 14 skills like setup, lint, prep, and code generation. It reads and enforces team conventions from a shared configuration file, supports design tokens and flow arrows, and integrates with front-end code and QA workflows.

Gates is a Claude Code plugin providing five stateless pre- and post-commit guardrails: it blocks git commits that change code without covering docs, workflow scripts with no per-agent model, agent dispatches without explicit model, unparseable config files, and nudges toward docs consolidation when drift is detected. Each gate has an acknowledgment marker for override and can be individually disabled via GATES_DISABLE in settings.json.

A German legal assistant for SMBs covering 12 practice areas of federal and state law with 16+ interactive calculators for employment, tax, data protection, and contract review. The plugin integrates with Google Calendar, Gmail, Microsoft 365, Slack, Box, Atlassian, and Egnyte to help teams manage legal workflows and compliance.

Hafıza adds persistent LLM-wiki memory to Claude Code, ingesting raw sources into interlinked markdown pages indexed for efficient querying without re-scanning. It includes an archiver agent for page synthesis, a lint agent for consistency checking, and a hook that protects raw sources from accidental modification.

A Claude Code plugin for writing product requirements documents, drafting and filing engineering tasks, and maintaining work logs—all configured to read from and write to your choice of markdown files, git repos, Notion, or custom trackers. It keeps product specs and task tickets synchronized, extracting work context from chat threads and generating accomplishment statements from daily logs.

This standalone Agent Skill repo provides two skills for working with SAP's UI Theme Designer and design tokens: ui-theme-designer-help answers how-to and conceptual questions about setting up, managing, and using UI theme designer on SAP BTP, while ui-theme-designer-design-tokens answers questions about SAP Design System themes, Fiori design token values across theme variants, and which tokens UI5/Web Components/Fundamental Styles controls consume.

AI Architect MCP Spec is a TypeScript MCP server that deterministically validates feature descriptions into verified 9-file PRDs by cross-checking symbols against a live codebase graph and ensuring every acceptance criterion traces back to its requirement. It runs on your machine with multi-judge consensus and Bayesian reliability calibration, catching hallucinated specs before code generation consumes them.

docs-index-keeper automatically keeps a repository's Markdown documentation index synchronized by running in pre-commit hooks or CI pipelines. It detects new or changed docs and updates the index table with zero configuration for standard setups and no external dependencies.

LoreDocs provides Claude with persistent access to organized, searchable project documentation through an MCP server and bundled skills, maintaining a SQLite-backed knowledge base with version history and full-text search. It integrates with Claude Code, Cursor, and other AI tools to inject relevant context into conversations and prevent knowledge drift across multi-agent environments.

Mercator AI maps codebases of any size using parallel Sonnet subagents and generates architecture documentation in docs/CODEBASE_MAP.md with a Merkle-tree manifest for O(1) change detection. A post-commit hook automatically refreshes file hashes in ~2 seconds with zero API tokens, enabling incremental updates of only changed modules without git dependency.

miadi-stateloom is a Claude Code plugin that bundles the stateloom design loom—an MCP server and 8 integrated skills—to let agents design system data (ERDs), behaviors (state machines), and usage scenarios (sequences) as a single coherent model. It automatically validates cross-model consistency, replays scenarios through machines, reconciles updates in propose or auto mode, and generates RISE rispecs from the complete system design.

mnema integrates a project's committed decision record and machine-local notes into Claude Code sessions automatically via SessionStart hooks, and connects an MCP server to enable the agent to read additional context and record its own work during editing. It also provides PreToolUse hooks to surface file-specific rules before writes and session tallies to track decisions and file changes.

This Claude Code plugin manages an Obsidian knowledge vault with skills for organizing files, renaming ambiguous content, fixing broken links, and capturing session insights as permanent notes. It includes protective hooks that prevent accidental writes to sensitive directories, published files, and redirect operational knowledge to standardized vault locations rather than per-project auto-memory.

Spec First guides developers through a three-step workflow: define requirements in an approved spec file, implement code from that spec by copying patterns from the repo, and generate documentation—all from the terminal. The plugin enforces spec-first discipline by requiring approval of requirements before implementation begins, with validation checks that gate each phase.

This plugin audits and fixes accessibility (WCAG 2.2) and internationalization issues across React, Vue, Svelte, Angular, and HTML, applying cross-element ARIA remediation, keyboard-focus corrections, color-contrast rewrites, hardcoded-string extraction with stable ICU MessageFormat keys, and RTL-ready CSS transformations. It works entirely offline on local files as reviewable diffs, never machine-translates or fabricates content, and cites the specific WCAG success criteria behind every change.

adr-creator helps developers document architectural decisions by generating Architecture Decision Records (ADRs) through Claude Code. The plugin includes one skill and one MCP server, requires npx, and supports tiers from free (50 req/day) to enterprise plans.

Automatically syncs AGENTS.md files from your repository into Claude Code's path-specific rules system on each session start, respecting .gitignore and maintaining a `.claude/rules/agents/` directory. Eliminates manual setup by discovering all AGENTS.md files and mapping each to the appropriate file scope based on its location in the repo.

alps-writer guides you through authoring ALPS (Application Layer Profile Specification) product requirement documents section-by-section via an MCP server in Claude Code. It integrates with the adr-writer plugin to hand off features as architecture decision records using the /feature-to-adr command.

This Claude Code plugin analyzes AI-generated Korean documents to detect and fix awkward phrasing, including compressed compound words, metaphorical expressions, and literal translation patterns, suggesting natural alternatives. It runs as a SessionStart hook with an accompanying shell script for rule validation.

Concise is a Claude Code plugin that intercepts verbose comments, oversized files, and padded PR/issue descriptions before they're written, offering pluggable em dash and AI writing pattern detection with presets for different teams. The agent reads denial reasons and rewrites flagged content in real time, preventing bloated output from reaching your working tree or GitHub.

The dc plugin is a brain round trip toolkit that decodes messy input, stress-tests plans, distills insights, and writes clean markdown output. It includes one agent, six skills, and a session-start hook to streamline analysis and documentation workflows.

This Claude Code plugin converts HWP (Korean word processor) documents to Markdown, either automatically when you ask Claude to read an HWP file or through explicit invocation of the `/docpler:hwp-convert` skill. The plugin downloads a platform-specific binary during installation and uses `docpler convert` to parse HWP files into readable Markdown format.

Hypermnesia MCP Viz is a read-only visualization layer that renders Claude Code sessions, Cortex memory stores, and codebase graphs as six interactive views: a project galaxy, 3D brain map, execution trace, consolidation kanban, knowledge browser, and wiki. It bridges Cortex's PostgreSQL database and local session archives to provide live reading angles without ever writing memories.

This Claude Code plugin equips Claude with comprehensive knowledge of Jamdesk's 25+ MDX components, docs.json configuration schema, 8 navigation patterns, CLI commands, and documentation best practices. It includes a quick-reference hub and on-demand detailed references plus an MCP server integration for querying Jamdesk documentation directly.

KolmoPDF converts PDFs to Markdown with layout preservation and handles formulas, tables, and multi-column layouts that standard tools struggle with, plus bidirectional conversion between Markdown and DOCX/HTML/PDF/LaTeX formats. The plugin includes cost estimation and credit balance checking, operating via an API key at kolmopdf.com with free tier and pay-as-you-go options.

A memory lifecycle system for Claude Code that adds save gates, tiered expiry (short memories archived when they lapse), consolidation passes, and a budgeted habit frame (HABITS.md capped at 8000 bytes/24 rules) to prevent hallucinated memories and knowledge loss across sessions. Includes spaced-repetition tutoring over learned habits and optional identity setup with user/assistant names, running entirely local with no external API dependencies.

NuBerea is an AI-powered theological assistant that provides scriptural research tools including Greek and Hebrew lexicon lookups, morphological analysis, cross-references, and textual criticism via an MCP server that integrates with Claude and other AI clients. The plugin enables word studies, exegesis, sermon preparation, and manuscript research directly within Claude Code.

PerMark is a free, open-source Markdown IDE that integrates with Claude Code via MCP to help you read and review what AI agents write, with automatic hooks that detect changes and enable document feedback. The plugin connects Claude Code to PerMark's desktop app, allowing you to watch agent edits in real time, leave comments on specific passages, and let agents read your feedback to steer their work.

This Claude Code plugin transforms Spotify podcasts you've listened to into a structured wiki organized by topic, with every claim cited back to its episode and source. It fetches episode transcripts via published text, local Whisper transcription, or fallback descriptions, then uses Claude to synthesize concepts across episodes and maintain a Karpathy-style knowledge base with disagreements tracked side-by-side.

The project-context-manager plugin scans README.md and architecture documentation to extract and enforce project conventions and architectural rules on every write operation in Claude Code. It provides a /project-context skill for manual rule extraction, an architecture-reviewer agent for deep structural analysis, and automatically enforces compliance via a SessionStart hook.

This plugin automatically activates at session start to enforce plain-language writing conventions across agent responses and documents, prioritizing clarity through answer-first structures, single ideas per sentence, and elimination of undefined jargon. It uses a session-start hook and shell integration to apply these formatting rules throughout the session.

Arkitect generates and edits native Draw.io and Excalidraw diagrams from JSON specifications, creating cloud architectures, system designs, data pipelines, and network views with 6,000+ bundled icons entirely on your machine. The agent validates diagrams and renders local previews, supporting both formal architecture visualizations and hand-drawn sketches with no external dependencies or uploaded data.

Cartographer is an agent-first code understanding system that builds a persistent world-model of your codebase with evidence-grounded facts, semantic maps, behavior flows, and navigable perspectives via semantic zoom. The plugin provides PR review overlays and persistent depth navigation through code understanding, powered by agents and skills that integrate with your development workflow via shell execution and session context.

This plugin scaffolds, validates, and generates marketplace presentation manifests for Claude Code plugins using native plugin files and optional presentation overlays. The marketplace-flow skill automates the entire workflow from bootstrap through validation to CI wiring, while individual commands handle initialization, migration, validation, and manifest generation.

This Claude Code plugin automatically exports conversations as readable text files whenever the context window compacts or a session ends, preserving full conversation history with formatted output. Exports are saved as timestamped files in the project's .claude/chat-exports/ directory and include user messages, Claude responses, and tool usage summaries while excluding raw JSON and system metadata.

A Claude Code plugin that conducts conversational daily journaling interviews using a 34-floor emotional altitude scale to identify your emotional state, optionally tracks behavior patterns, and saves structured Obsidian markdown entries. It features optional advisory panel commentary from named thinkers and can integrate with RescueTime for grounded accountability based on productivity data.

This Claude Code plugin scans recent sessions, journals, and decisions to detect recurring patterns, then captures them as CLAUDE.md rules, concept notes, writing seeds, or skill improvements. It runs on-demand via `/patterns` or semi-autonomously at session end with targeted suggestions based on tool friction, corrections, dead-end recovery, and discoveries.

A four-skill Claude Code plugin for converting work outputs into external-facing deliverables: deliver-report writes messages/emails from completed files, test-report-docx generates signed-off test reports, daily-report summarizes daily work across projects and sends via Gmail OAuth/SMTP/MCP, and to-questionnaire turns decisions into third-party survey markdown. All outputs share strict rules prohibiting AI/internal-process/revision language and assume non-technical readers.

The docs-generator plugin creates academic and technical documentation including writeups with IEEE citations, technical specifications (ADR/RFC/design docs), and structured reports, with optional PDF export via pandoc. It provides three skills for generating different document types, bundled templates, citation styles (IEEE, APA, Chicago), and a SessionStart hook to warn about missing PDF dependencies.

docs-guardian flags when code changes likely make docs, README, or API references stale and proposes specific edits needed. It maps changed source files to affected documentation, verifies genuine mismatches, and suggests surgical updates via a `/docs-check` command and PostToolUse hook.

docs-workflows scaffolds documentation repositories with Material for MkDocs, handles coverage audits and backlog prioritization, and synthesizes product documentation from PRDs and implementation diffs with style-checking and Opus review gates. It provides seven commands for documentation lifecycle management: initialization, branding, serving, auditing, profiling, and release-note generation.

dogovornoe-pravo is a contract review plugin for Russian legal compliance that verifies client agreements against Russian Civil Code norms covering contract formation, liability, warranties, and subscription models, while also triaging confidentiality agreements and tracking contract revisions. It integrates with Google Drive to manage contract documents and provides escalation routing and business-friendly review summaries for stakeholders outside the legal function.

This engineering plugin provides 10 skills for software development workflows: git/PR helpers (commit, branching, pull requests), code-review and manual-testing guide generation, system design workshopping, ADR recording, bug investigation, and project setup orchestration. It includes 4 role-based sub-agents (code-explore, implementer, heavy-implementer, test-runner) that delegate tasks to cost-appropriate Claude models, with optional role-based delegation criteria injection.

foreman is a Claude Code plugin that orchestrates feature development by having a main agent (foreman) review requirements, write plans, and delegate tasks to a subagent worker, then verify results and track progress. It manages the entire workflow from feature definition through contract approval, task execution, and documentation in a workbench folder.

This Claude Code plugin helps writers produce clearer, more accessible prose by replacing em-dashes and similar punctuation with plain-language connectors in documentation, READMEs, and commit messages. It operates as a prose-specific formatter that activates at the task edge without steering mid-solve, ensuring readable English for native and non-native readers alike.

A Claude Code plugin providing German-language financial accounting and compliance workflows tailored to HGB and EÜR standards for GmbH, GbR, sole proprietorships, and freelancers. It integrates with MCP servers for ERP, payroll, and data warehouse systems, and includes specialized skills for VAT reporting (USt-Voranmeldung), eBilanz submission, payroll processing, and GoBD compliance.

This plugin provides compliance and analysis tools for Russian government procurement under two distinct legal frameworks: the 44-FZ contracting system for federal purchases and the 223-FZ regime for corporate purchases with state participation. It serves both buyers and suppliers with features including eligibility verification, documentation analysis for bias toward specific vendors, bid composition assistance, FAS complaint filing, and counterparty verification against separate lists of unreliable suppliers.

Humanizer removes AI-generated writing patterns from text by analyzing against Wikipedia's Signs of AI Writing guide and optionally calibrating to the author's personal voice fingerprint. It detects 29+ patterns including signature phrases, formulaic structures, and style markers, then rewrites the text with a final audit pass to catch lingering AI characteristics.

A Legal Lifecycle Framework Claude Code plugin with 35 skills that enable assessment, drafting, review, compliance, and governance workflows for legal operations. It integrates via a PreToolUse hook and phase-gate.sh external tool, requiring no third-party API keys.

This plugin registers the context7 MCP server, which provides up-to-date library and framework documentation lookup so Claude's responses reflect current docs rather than training data. It requires only Node.js and runs via a platform-agnostic launcher script, with optional global installation of @upstash/context7-mcp to improve startup speed.

A Claude Code plugin that advises on Russian tax law, helping select optimal tax systems (OSNO/USN/AUSN/PSN/NPD/ESHN) and analyzes VAT treatment under 2025–2026 reforms, VAT analysis (rates, exemptions, deductions, refunds), tax audit responses, FSN correspondence, pre-trial appeals, and transfer pricing. The plugin provides draft analyses for verification by practicing tax consultants and lawyers, with outputs intended as starting points rather than final legal advice.

A Claude Code plugin providing specialized analysis of Russian real estate law and contracts, including shared construction agreements (ДДУ under Law 214-ФЗ), ready property sales, commercial and residential leases, and cadastral/EGРN property verification with encumbrance analysis. Each output is a draft document designed for review by a practicing attorney.

This Claude Code plugin compiles knowledge from Obsidian vault sources (meetings, daily notes, projects) into a persistent, cross-referenced wiki at wiki/ with ten slash commands (/wiki-ingest, /wiki-query, /wiki-save, etc.) and integrated task management via Obsidian Tasks. It supports both Claude Code and GitHub Copilot as first-class agents with session continuity via hot.md caching and hooks.

This Claude Code plugin distributes OMEGA framework knowledge as 14 skills, 8 hooks, and 1 MCP server that stay version-matched with the monorepo code they document. It automatically injects context-aware framework skills based on detected @omega.js packages in the working project, with mechanisms to support both published and local development copies.

ParaQualis-GXP is a Claude Code plugin for computer-system validation and life-sciences regulatory compliance, covering GAMP 5, 21 CFR Part 11, EU GMP Annex 11, live eCFR data, and openFDA APIs with an IQ/OQ/PQ qualification-pack generator. It provides 18 slash commands, 2 auto-invoked advisor skills, 3 specialist sub-agents, and includes an MCP server for openFDA data access (free API key provisioning via `/openfda:setup`).

This plugin implements Russian data protection compliance checks according to Federal Law 152-FZ, providing triage for data processing (consent, data localization, harm assessment), handler authorization review, subject rights responses, and gap analysis against regulations. It integrates with Google Drive via MCP to monitor and document data processing policies and compliance artifacts.

A fourteen-command pipeline for authoring Product Requirements Documents and related artifacts through role-based workflows spanning idea refinement, grounding, customer review, and epic breakdown. Each stage gates on an Opus-backed review agent, with an optional six-command BRD-to-PRD intake route for grounding customer requirements documents.

speak-human-en enforces clear communication patterns in Claude responses through pre-question self-checks and structured expression rules (S1–S6), derived from 548 real decision records about user questioning. It applies automatically on SessionStart and can be pinned always-on via a local config file to ensure responses verify facts, brief context, gloss jargon, and report progress or blockers in a Done/Not done/Known issues format.

This plugin connects Claude Code to the SpecsGraph MCP server to offer guided domain modeling workflows across four specialized skills: director (routes to the right specialist), engineer (models domains and bounded contexts), product (writes features and scenarios), and brownfield (maps existing codebases). All workflows stage changes as proposals for team review before publishing to the specification model.

This plugin enforces TigerBeetle's Tiger Style coding methodology for Zig projects, providing assertions, bounded loops, and strict formatting rules within Claude Code sessions. It auto-detects Zig projects, injects style guidance into CLAUDE.md, and offers commands to reference patterns and audit files for mechanical violations.

trustmebro helps you manage plan-first projects by maintaining three synchronized Markdown files: LAYOUT.md for code structure, PLAN.md for development planning, and ROADMAP.md for progress tracking. The plugin uses Python scripts with lifecycle hooks to keep these documents current as you work, ensuring your project organization stays aligned with actual implementation.

A documentation-driven multi-agent development workflow plugin with three-layer architecture (orchestration/execution/review) that auto-generates project scaffolds, decision tracking, and business SOPs with screenshots for deployed web applications. It uses session hooks and signal-word detection to capture decisions in CLAUDE.md, enforce pre-commit gates, and coordinate team reviews with built-in risk management discipline.

This Claude Code plugin intercepts user input via the /wtf command to provide dense, visually-structured explanations of preceding messages, replacing verbose descriptions with ASCII diagrams, flowcharts, and tables. It enforces a "explain-only, never execute" principle through PreToolUse guards and integrates with the archify tool to generate optimized workflow visualizations for complex relationships and processes.

This plugin wraps the @upstash/context7-mcp server to provide live, version-accurate library documentation lookups for Claude Code bundles. By consolidating the Context7 server into a single shared plugin, multiple bundles can depend on one definition without triggering MCP deduplication warnings.

This Claude Code plugin injects corrective Zig syntax patterns into your project context to fix Claude's outdated training data, auto-detecting whether your project targets Zig 0.15.x or 0.16.x and applying the matching reference covering 25 major breaking changes. It runs at session start to populate CLAUDE.md with version-aware corrections for Writergate, I/O interfaces, namespace moves, API renames, and container changes.

An MCP server that guides users through creating Architecture Decision Records (ADRs) in MADR format with interactive Q&A workflows and optional PRD traceability. It provides tools for ADR lifecycle management, document storage, and structured conversation guides across all key decision sections.

Agent Ergonomics is a reusable skill that reconstructs systems from an operator's perspective to identify friction and make coordinated changes to specifications, documentation, and design, enabling AI agents to understand, control, verify, and recover within systems reliably. It provides a complete workflow with reference materials, optional validation tooling, and behavioral evaluation fixtures to improve system design without substituting prose for implementation evidence.

befund performs comprehensive static analysis of live codebases—auditing architecture, documentation drift, CI/CD topology, conventions, code idioms, business rules, and UI accessibility—with every finding tied to file:line evidence and gated transformations recorded on disk. It pipes findings through a read-only CHECK phase and gates any source modifications behind explicit human authorization in a FIX phase, distinguishing it from design conformance tools and architectural codifiers like lehre and cupertino.

Converts AI learning conversations into illustrated study guides on the Bloom platform, and enables searching existing guides. The plugin integrates with bloombook.cn to streamline educational content creation from chat interactions.

Canon is a Claude Code plugin that enforces a structured spec-to-production workflow using skills and git hooks to manage specs, code reviews, documentation, and deployments. It includes 10 skills (define, approve, build, review, canon, staging, production, ship, and setup commands) that guide work through branches with validation at each stage, plus 10 hooks that enforce git rules around the workflow.

Claude Code plugin that injects contextual educational insight types (Curious, Insight, Flow, Tradeoff, Pattern, Level Up, Concept) into responses at code-appropriate moments, with a persistent personal glossary that grows over time. The Curious insight uniquely triggers before execution to question assumptions, while others appear after to explain implementation decisions, execution paths, design tradeoffs, patterns, transferable knowledge, and terminology.

A Claude Code plugin that turns an Obsidian vault into a provenance-tracked knowledge base following Andrej Karpathy's LLM Wiki pattern, with an eight-agent orchestrator stack and 18 hooks enforcing schema at every tool-call boundary. The system handles ingestion, querying, linting, synthesis, and maintenance through skill-based commands, using the Bun engine to verify, fix, and checkpoint every auto-heal with git reversibility.

coldwrite is a pre-Write hook that uses a stateless reader model (without conversation history) to check prose documents for comprehensibility before writing, blocking commits until unclear passages are fixed. It targets only prose file extensions (md, mdx, adoc, asciidoc, rst, txt) and skips content under ~300 characters or non-prose fragments, returning detailed error reasons to enable iterative correction within the same turn.

Automatically saves every Claude Code session as a human-readable Markdown file, organizing logs by project with timestamps and extended thinking preserved as collapsible blocks. The plugin runs on the Stop hook after each Claude response and provides a permanent, readable archive that persists beyond Claude Code's default 30-day cleanup.

Curt is a Claude Code plugin that provides editorial guidance and prose linting to help refine writing while preserving meaning and author voice. It offers targeted editing suggestions through two skills—anti-slop for prose and anti-slop-code for code comments—and integrates with session hooks to inject editing rules at the start of work.

docex is an MCP server that lets you edit Word documents (.docx) programmatically with code instead of menus, and converts them to/from a plain-text .dex format that preserves all formatting, tracked changes, and comments. It includes CLI commands and a JavaScript API for academic manuscript workflows like responding to peer review, applying journal formatting, comparing versions, and validating submission requirements.

This Claude Code plugin enforces a strict six-phase waterfall SDLC methodology with doc-first practices, TDD validation per task, and a two-pass reconciliation gate that blocks phase sign-off until code and documentation align. It injects an always-on 5-rule contract at session start, provides on-demand workflow guidance via the `/sdlc-strict-waterfall` skill, and measures token cost and quality metrics per phase to keep AI coding agents accountable to signed specifications.

This standalone Agent Skill provides access to four MCP servers for EU AI Act and multi-regulation compliance, including real-time EUR-Lex full-text search across six EU regulations (EU AI Act, DORA, NIS2, Cyber Resilience Act, CSRD, GDPR) and compliance auditing tools. It enables AI agents to classify AI systems for regulatory risk, generate signed compliance artifacts, search regulation text with FTS5, and track enforcement deadlines—all with optional cryptographic attestation for audit defensibility.

feynman automatically injects ASCII diagram rules into Claude Code, Codex, and OpenCode prompts at session start, transforming prose descriptions of flows, hierarchies, comparisons, and status into visual diagrams. The plugin survives context compaction by re-injecting rules after `/compact` and `/clear` commands without per-turn overhead.

This plugin monitors AI citations and optimizes content for visibility across ChatGPT, Perplexity, and Google AI Overviews using a 5-layer GEO Stack scoring system. It includes citation monitoring, crawl parity analysis, structured data validation, and AI-optimized content strategy tools powered by DataForSEO, Google Search Console, and Firecrawl integrations.

html2wp converts static HTML sites (from Lovable, Bolt, v0, Next export, etc.) into standalone WordPress block themes with pixel-perfect verification, pages, blog, forms, menus, SEO and redirects—all without requiring any plugins. The conversion includes multiple verification gates to catch fidelity drift and ends with a manual page-by-page review before delivery.

This plugin detects Korean-language AI writing patterns in past Claude Code responses using the im-not-ai taxonomy (85 patterns) and automatically injects a personalized style guide into new sessions to reduce detectable AI tone. It re-analyzes accumulated sessions weekly and maintains the top 15 patterns, tracking evicted patterns for potential re-inclusion.

ido4specs analyzes your codebase and decomposes strategic specifications into structured technical specs with effort, risk, and dependency metadata for implementation planning. It bridges high-level requirements to actionable GitHub issues through a methodology-neutral task breakdown pipeline.

This plugin enforces a structured reply format for Claude Code agents using ASD-STE100 principles: key-first bullet points, a one-sentence conclusion, and eight fixed status sections (Goals, Done, Doing, Todos, Pending, Quests, Risks, Ideas) that remain visible even when empty. The plugin intercepts session start and user prompt submission to consistently format agent responses in any language, making critical information (failures, blockers, approvals needed) scannable in seconds.

lockwire fingerprints code symbols across four separate tiers (path, signature, body, dependencies) and binds documentation claims only to the tiers they depend on, catching stale CLAUDE.md, AGENTS.md, and docs at write time with tamper-evident history. It runs as a Claude Code plugin with pre-commit and hook integration, using tree-sitter and BLAKE3 to detect documentation drift without false positives from internal refactors.

LSA (Living Spec Architecture) is a technology-agnostic spec layer that grounds requirements and acceptance criteria in your codebase before building, then verifies the delivered diff afterward to ensure it does exactly what was specified. It wraps any coding agent (Claude Code, Cursor, human) with industry-standard EARS requirements and Gherkin scenarios, performing two critical checks: verify (before) to ensure specs reference real, buildable code, and reconcile (after) to confirm the diff matches the spec's intent.

Matrize derives a master design system from reference exemplars, assigns explicit purpose and rules to every recurring element, and emits to platform-neutral DTCG token files that generate byte-stable output across CSS, Tailwind, VitePress, and other targets. Measurement and interpretation are kept as separate artifacts; sources are read-only, and values are graded for trustworthiness and reproducibility rights before any token is created.

An MCP server plugin that provides search and retrieval capabilities for official AWS documentation, including API specifications, best practices, and service-specific guides. It integrates with Claude Code and other platforms to enable direct access to AWS docs without leaving the development environment.

This MCP server plugin wraps Microsoft MarkItDown to convert various document formats (PDF, Office, images, HTML, CSV, JSON, XML, etc.) to Markdown from HTTP/HTTPS URLs, local file paths, or Data URIs. It provides a `convert_to_markdown` tool for use within Claude Code and integrates with the `ndf:scanner` agent for document processing workflows.

The memory plugin captures session records deterministically via a SessionEnd hook and distills them into classified vault notes on demand through a two-phase-review skill. It provides only the archival mechanism—no built-in identity content, personas, or communication channels—leaving those to be defined in your own vault.

mif-docs ships 47 skills, one per concrete document genre (tutorials, how-tos, references, architecture docs, ADRs, runbooks, PRDs, and 30+ more), each adopting industry patterns while guaranteeing MIF (Modeled Information Format) compliance through deterministic validation against a canonical JSON-LD schema. Every output document is both human-readable in its native genre and machine-readable MIF, with a fail-closed guard that blocks non-conformant output before it's written.

A Claude Code skill that reformats agent responses for TDAH readers by prioritizing the answer/action on the first line and applying strict formatting rules (numbered steps, immediate term translations, line limits). Activates automatically on SessionStart and supports two modes: EXECUTAR (execution-focused with concrete state) and ENTENDER (understanding-focused with one topic at a time).

Myelinate is an open-source skill suite that provides university-standard tutoring, courses, labs, assignments, tests, flashcards, and study plans for learning concepts and codebases. It routes requests to specialized skills that teach step-by-step with rendered visuals, honest feedback, and no agreeable shortcuts.

okl (Observed Knowledge Ledger) is a Claude Code plugin that briefs coding agents with contextual lessons—engineering rules, architecture decisions, and verified documentation—before each task via MCP hooks, and records what was learned to keep knowledge persistent and drift-checked. It requires the okl CLI tool and integrates via pre-task and end-of-session hooks to maintain a shared, verifiable knowledge store across team repositories.

plainspeak translates Claude Code's technical explanations into plain language for non-developers building software intuitively, hiding code paths and jargon while preserving actionable next steps. Users control verbosity (lite/full/strict) and voice style (tone, formality, length, formatting) via simple commands that persist across sessions.

prime-context is a Claude Code plugin that manages project memory through a thin AGENTS.md router and on-demand context files in `.claude/context/`. It captures session learnings, audits structure drift, and runs session-start verification to ensure memory coherence.

Maintains a self-healing directed acyclic graph of LLM-authored wiki pages inside Obsidian that automatically syncs with your codebase using PostToolUse hooks. The plugin ingests sources (Prisma models, API routes, user stories, EIS documents), synthesizes structured wiki pages with bidirectional cross-links, and provides commands to audit the knowledge graph, explore features, and author implementation specs.

A multi-agent orchestration pipeline for Qlik Sense that automates the full development lifecycle from requirements through documentation, producing production-grade load scripts, expressions, visualizations, QA reports, and comprehensive technical artifacts across a structured 9-phase workflow. The pipeline pauses at approval gates and execution validation checkpoints to incorporate your feedback and live reload testing.

Rupa is a Claude Code plugin that converts Markdown to HTML or PDF, keeping rendering markup out of Claude's token output. It includes a skill and a SessionStart hook to enable Markdown-to-rendering workflow directly within Claude Code.

Signal is a Claude Code plugin that enforces plain-English writing standards by stating token budgets before an agent writes and scoring the reply against thresholds for verbosity and clarity. It uses hooks and skills to block overly long or complex responses, optionally forcing rewrites before showing the result to users.

skill-zh-cn automatically translates skill descriptions, commands, and sub-agent metadata from English to Simplified Chinese across multiple hosts (Claude Code, zcode, opencode, Codex, DSH) using the host's own agent—no external APIs or CLI dependencies required. The plugin operates via a SessionStart hook that scans for untranslated content and applies line-level patches, preserving original English in a backup field for full reversibility.

ste-mode makes Claude write in Simplified Technical English style with short active sentences and plain words, toggled via the /ste command. It injects writing rules at session start to produce clearer, more readable technical explanations.

One-command bootstrapper that installs Obsidian, the claude-obsidian LLM Wiki plugin (from TechTrip's maintained fork), and ships ingest skills for YouTube, X/Twitter, voice, code, and NotebookLM sources plus a new-idea scaffolder, with git sync and backup pre-configured for a fresh Mac. Runs interactively and idempotently, pre-checking and diagnosing the system with included doctor and repair tools.

Routes text-heavy documentation tasks (planning, PRDs, READMEs, CLAUDE.md) to leaner Claude models like Haiku and Sonnet to save tokens while maintaining quality, with auto-detection of task keywords and explicit model-tier commands. The plugin returns results to your original session model once the task completes.

The Trade Safety plugin provides an MCP server that explains the Trade Safety sandbox's capabilities, input limits, and execution boundaries. It connects to the LiquiLens Trade Safety service without requiring a product API key, though users may need their own account or plan.

unwordy enforces team writing conventions for coding agents across commits, pull requests, code reviews, and design handoffs using configurable profiles and local pattern checks. The plugin preserves factual content and repository conventions while flagging filler and verbose language, with hook integration for real-time enforcement and standalone checking via shell commands.

This Claude Code plugin automatically archives conversation sessions into an Obsidian-compatible wiki, maintaining both a global vault and per-project documentation vaults. At session start, it injects relevant vault context—including past decisions and recent sessions—so Claude can reference prior work without rediscovering it.

volens maintains an append-only decision log and automatically derives a design snapshot from it, keeping project documentation in sync with architectural decisions without manual updates. The plugin runs hooks at session start and stop to incrementally refresh the design doc whenever decisions are logged.

Accents for UI5 is an open design framework for building business applications on OpenUI5, providing a shell, 16 page patterns, and 70 pre-built elements with consistent design rules for planning, analytics, forms, reporting, and workflow. The Claude Code plugin includes a skill that teaches the framework's rules and an MCP server that helps find entries, read examples, launch apps, and validate implementations.

This standalone Agent Skill provides two skills—acuris-address for address validation, geocoding, and reverse geocoding using the Acuris APIs, and acuris-eudi for EUDI Wallet verification as an OID4VP/SD-JWT VC relying party. The skill teaches API surface details, SDK usage patterns with @acuris-geo/av-sdk and @acuris-geo/centra-checkout, implementation recipes for checkout flows and migrations from competing vendors, and EU 27 trust validation for KYC workflows.

agent-docs-nextra provides guardrails for Nextra documentation sites, including a generated-guard hook that blocks hand edits to generated pages and agents for security and SEO validation. It installs rules, lint configs, quality-gate checks, and pull-request CI setup to prevent common docs issues like secrets in exports and missing metadata.

This plugin manages AI documentation for Claude Code projects through a lifecycle of bootstrap, lookup, update, check, and audit operations. It maintains a docs-ai/ directory with verified-against stamps, automatically reminds developers to check documentation before code changes, and provides skills for initialization, updates, staleness checking, and comprehensive audits.

Almagest is a Claude Code plugin that pairs with a standalone binary tool to build and maintain an LLM-powered wiki within Obsidian, with agents ingesting sources, linking repositories, and proposing changes that you approve before they enter the vault. It provides three read-only agents, fourteen skills, and features like document checkout, journaling, and safe deletion to keep your knowledge base human-navigable while agents handle the heavy lifting.

An Agent Skill that teaches AI coding assistants how to correctly set up Angular SSR, diagnose NG05xx hydration errors with a structured 7-step process, and apply correct SSR patterns in Angular v17+ applications. It includes an anti-pattern blacklist to prevent common mistakes like misusing ngSkipHydration or typeof window checks.

Artifact Hub automatically saves every Claude artifact you publish to a local JSON file and displays them in an organized folder-based index page called Project Hub, which you can search and reorganize by asking Claude in plain language. The plugin runs a Node.js hook on artifact publication to track links and titles, then lets Claude file pages into nested project folders while keeping all links safe locally.

AsItIs provides read-only MCP tools and a skill to safely read, lint, and analyze markdown and agent steering files while teaching Claude minimal-edit practices and optionally guarding against accidental file corruption. The plugin enforces byte-level fidelity—preserving line endings, trailing spaces, BOMs, and frontmatter—through an MCP server with no network access, no write capability, and zero external dependencies.

atlas-obsidian pairs an MCP-served Go binary with a Claude Code plugin to manage project context as a source-cited Obsidian wiki and versioned work threads that move through stub, spec, plan, and receipt stages. It integrates with project workflows through lifecycle hooks, ingesting external sources into tracked Markdown while keeping work state queryable and auditable across agent sessions.

Babel Fish auto-generates a 19-section project map and vocabulary translation layer that gives Claude instant, accurate knowledge of every route, model, service, feature, and infrastructure element in a codebase. The plugin introspects the full stack, self-updates via pre-commit hook, and pairs optionally with semantic-memory to enable deterministic code-to-path translation.

A Claude Code plugin that wraps the Backlog.md MCP server, auto-installs the backlog-md CLI on first use, and provides a `/backlog-migrate` skill to convert messy `/docs` folders into a clean task backlog. Three gated hooks automatically track plan acceptance and git commits while staying dormant in projects without a backlog/ folder.

Bayan helps Claude write clearer, more natural text in English and Arabic by automatically removing machine-like habits (filler phrases, hidden characters, AI signatures) from files and providing line-by-line readability feedback. It adjusts writing style based on reader level (no-code, junior, or developer) and works across all output from the Nexika plugin suite.

A security toolkit for brain-computer interface (BCI) projects that performs threat modeling, vulnerability scanning, and neuroethics compliance checking entirely locally without external API calls. It includes 8 tools for scanning BCI configurations, detecting patterns, scoring device vulnerabilities with NISS scoring, and validating regulatory and neuroethics compliance against published guardrails.

This Claude Code plugin provides a `bmc-spec` skill that answers technical questions about BMC firmware specifications (IPMI, DMTF MCTP/PLDM/SPDM, Redfish, NVMe-MI, OCP, and others) and OpenBMC source code with full document citations. It maintains a local library of 113 specification documents and repositories, automatically fetches missing versions from publishers or the Internet Archive, and cites exact document sections and page numbers in every answer.

Brain distills Claude Code and Codex sessions into a persistent private wiki that compiles digests and external sources into entity and topic pages, injecting key facts as a brief into each new session to build compounding context over time. The plugin hooks into session lifecycle events to automatically digest conversations and manage contradictions through human-guided ingest passes.

The browncoat plugin provides guidance on general software design and development practices including box decomposition, interface contracts, error and retry handling, test boundaries, and clean-code conventions. It includes an agent with three supporting skills that activate on session start to help developers apply these architectural and code quality principles.

This plugin enforces consistent voice and quality standards for development documentation by combining writer personas, measurable quality gates, and an editorial review layer that catches tone drift before content ships. It produces seven specialized artifact types—ADRs, design docs, handoffs, changelogs, and end-user guides—each routed through composable personas (Technical Writer, Doc Writer, Marketing Copywriter, Context Curator) and validated against deterministic checks via the bito tooling.

This Claude Code plugin records rules and corrections you make into checked-in context files (AGENTS.md, design.md, CLAUDE.md) so they persist across sessions and teammates without repeating yourself. It uses a UserPromptSubmit hook to detect when you state a rule, routes it to the most specific applicable file, and optionally suggests grep-based audit checks to enforce the rule.

This plugin packages IBM's official carbon-builder skill with a hosted Carbon MCP server to enable component/variant/prop search, docs lookup, accessibility guidance, charts source, labs access, and code-compliance audits for IBM Carbon Design System UIs. An IBM-approved account is required to access the server; unapproved sessions fall back to a bundled public-docs index.

Chronicler generates standardized `.tech.md` documentation from your codebase using LLMs and tracks when docs drift from source files via merkle tree hashing. It stores documentation in version-controlled Markdown with YAML frontmatter, making it AI-readable and RAG-ready while keeping docs synchronized with code through freshness detection and stale-file reporting.

claude-intent records code decisions, design rationale, and alternatives considered during development, storing them alongside conversation transcripts for later retrieval and relationship tracking. It provides four skills to record decisions (intent-record), trace decision history (intent-why), refine earlier decisions with new context (intent-refine), and retract decisions (intent-retract).

Detects CLAUDE.md instruction files that Claude Code missed because Bash operations or cd commands bypassed the file tool triggers, and injects their content into context via the InstructionsLoaded hook. Tracks loaded instructions and watches shell commands to identify directories whose nested rules were silently skipped.

claude-obsidian is a Claude Code plugin that builds and maintains a persistent, compounding wiki vault in Obsidian using hybrid retrieval (contextual prefix, BM25, cosine reranking) and supports multiple organizational methodologies (LYT, PARA, Zettelkasten, Generic). It ships with 4 commands (/wiki, /autoresearch, /canvas, /think), 3 agents, 17 skills, per-file advisory locking for multi-writer safety, a pre-commit verifier agent, and an optional DragonScale Memory extension for semantic tiling and autoresearch.

claude-wiki writes, validates, and queries wikis based on code and documentation using Open Knowledge Format (OKF) with YAML frontmatter and markdown. It generates node drafts from source code, validates on save, maintains an index, detects contradictions, and answers questions grounded in wiki content.

cleat centralizes repository agent instructions in a single AGENTS.md file that all agents read, reducing CLAUDE.md to a pointer via hooks that fire before and after tool use. The plugin enforces instruction consistency across agents through PreToolUse and PostToolUse hooks, with no resident commands or skills.

A Claude Code plugin that generates a machine-readable knowledge base in the CKB v0.2 format inside any repository at ckb/, including eight human-readable architectural docs (C4 views, workflows, business rules, security, operations) plus a structured ckb.json and provenance-tracked claims with confidence levels. It supports 30+ manifest types and 20+ web frameworks, enforces read-only writes to ckb/ via hooks, and skips regeneration when the knowledge base is fresh.

Codexpiator is a comprehensive Claude Code plugin that routes technical questions across nine specialized skills covering frontend, backend, security, testing, architecture, DevOps, Git/DX, and AI integration, with automatic detection of external specialized skills when available. The plugin includes three slash commands for audits and setup, enforces best practices through Git hooks (blocking AI attribution by default), and provides guidance organized as modular, single-purpose files rather than monolithic advice.

This plugin preserves commit rationale and context for future reference during development by storing meaningful background information about code changes. It hooks into session start to make commit rationale accessible on demand throughout the development workflow.

This plugin hooks into response completion to review each output for concision and clear technical English, applying a single correction pass with loop protection to prevent infinite refinement cycles. It operates as a post-processing filter that runs locally via Node.js without requiring external APIs.

The Context7 plugin bundles an MCP server into Claude Code that fetches current, version-specific library documentation and code examples on demand instead of relying on training-cutoff knowledge. It indexes thousands of libraries and frameworks, allowing Claude to query up-to-date docs and working code snippets directly from source material.

Context7 is an MCP server that fetches current documentation and code examples directly from source repositories to provide Claude Code with up-to-date API information, eliminating reliance on stale training data. It includes tools to search for libraries and their versions, then retrieve version-specific documentation ranked by relevance to user queries.

Context7 bundles an MCP server into Claude Code that fetches up-to-date, version-specific library documentation and code examples on demand instead of relying on the model's training cutoff knowledge. It indexes thousands of libraries and frameworks, allowing Claude to pull current docs and working code snippets directly from the source when queried.

A Spanish legal contract plugin for law firms that drafts, reviews with risk flagging, negotiates counterproposals, and handles enforcement procedures (demand letters, resolution, remedy claims, court filing) for civil and commercial contracts. It integrates with Jurisprudenciator for live legislation, case law with ECLI citations, commercial registry data, and property records verification.

This Spanish labor law advisory plugin helps law firms and social security graduates with employment contracts, dismissals, and compliance through 31 specialized skills that integrate with Jurisprudenciator for live legislation, case law with ECLIs, and citation verification. It covers contract modalities, payroll calculations, disciplinary action, dismissal letters, reconciliation filings, and litigation support—all conforming to Spain's Workers' Statute and collective bargaining agreements.

Coral is a Karpathy-style LLM wiki maintainer bundled as a Rust binary that scaffolds multi-repo documentation, manages dev environments, and exposes your codebase to Claude Code and other agents via Model Context Protocol. The Claude Code plugin registers five auto-invoked skills, two slash commands, a SessionStart hook, and an MCP server that lets Claude read your repo's wiki, manifest, lockfile, and test results with a modern WebUI featuring a force-directed graph and provider wizard.

The core foundation plugin provides Earth science data handling, statistics, uncertainty quantification, cartography, quality control, and reproducible analysis workflows for Claude Code. It connects to NASA's Earthdata repository and includes 12 skills spanning data formats (NetCDF, Zarr, GeoTIFF, GRIB), analysis workflows, and quality assurance tooling.

This plugin generates professional credit memorandums in Word, PDF, and PowerPoint with 10 embedded charts by automatically collecting 3–5 years of financial data from public sources (SEC EDGAR, SerpApi, Stock Analysis, Macrotrends, DiscoverCI) and optional premium providers. It supports both public and private companies, enforces 5-layer guardrails, and includes comprehensive financial, risk, repayment, industry, and management analysis.

This QA plugin scans Markdown client deliverables against a team rulebook, recomputing metrics and separating blockers from warnings while leaving the draft unmodified. It validates findings against rule definitions and metric formulas, then issues a ready-to-send verdict based on evidence and severity.

This skill instructs Claude to enforce a comprehensive set of production-grade development standards across the full SDLC — covering code quality, security, performance, API design, testing, DevOps, and documentation — across React, Next.js, Vue 3, Svelte, Python/FastAPI, Node.js/Express, Go/Fiber, PostgreSQL, and GitHub Actions. Users invoke it with `/dev-standards` followed by a task (e.g., code review, API design, CI/CD setup, schema design, testing) and Claude applies the relevant standards automatically without additional prompting.

This standalone Agent Skill repo provides two skills: one for generating README content and another for managing SEO metadata. The skills can be installed and used with the `npx skills add` command without requiring external API keys or third-party services.

This Claude Code plugin transforms PDFs into Markdown format through an agent, reducing operational costs by blocking direct PDF reads via a PreToolUse hook. The included block-pdf-read.sh script ensures large document sets are processed efficiently without expensive direct file access.

dq-session-db archives every Claude Code session locally in a searchable SQLite database with gzip-compressed transcripts, letting you ask Claude about past work and resume any session without external services or lock-in. It integrates via hooks at session start and end, stores everything in plain files you control, and optionally generates AI-written session summary cards using a model of your choice—local or cloud.

EasedUP is a K-12 education assistant that creates curriculum-aligned lesson plans, units of work, assessments, activities, and student-facing resources through explicit slash commands and embedded teaching strategies. It works standalone with user input and optionally integrates with Australian Curriculum v9, Common Core State Standards, and UK National Curriculum MCP servers for live standards lookup.

Englint is a Claude Code plugin that intercepts every user prompt to provide an English grammar and clarity check before executing the actual task, showing corrections, advanced alternatives, and grammar rules. It logs your writing over time and generates progress reports to help you improve your English while prompting Claude.

This Claude Code plugin provides a complete set of hooks that validate and render HTML pages (explanations, reports, judgments) written by agents through a standard renderer pipeline with quality assurance. It checks agent-generated pages with terminology hovers, diagrams, and evidence citations using hooks triggered at key workflow points (PostToolUse, Stop, SubagentStop, UserPromptSubmit).

This plugin recreates the deprecated Explanatory output style by injecting a SessionStart hook that encourages Claude to provide educational insights about implementation choices, codebase patterns, and design decisions alongside code work. It formats these insights in a distinctive box and activates automatically at the start of every session, adding token cost for the additional context and verbose output.

Plugin de Derecho de extranjería español que prepara solicitudes, memorias, alegaciones y recursos (arraigos, nacionalidad, protección internacional, expulsiones, etc.) conforme al Real Decreto 1155/2024 y Ley Orgánica 4/2000, integrándose con Jurisprudenciator para acceso a normativa vigente y jurisprudencia de tribunales españoles y europeos. Incluye 34 skills organizadas por áreas (entrada de caso, arraigos, residencia y trabajo, familias, nacionalidad, asilo, expulsión y recursos administrativos) y conectores opcionales para Google Drive, Microsoft 365, Dropbox, Box y DocuSign.

The fix-docstrings plugin audits and auto-fixes Python docstrings to Google style, checking for missing summaries, wrong section order, and undocumented parameters. A hook detects when LangChain tools explicitly enable parse_docstring and injects just-in-time constraints to prevent parser failures from malformed continuation lines.

A genealogical research workflow plugin for HTML report workshops that integrates web search, page extraction, image analysis, and optional MCP servers to organize family history investigations. The plugin provides six skills for research, resume, new reports, build chain, MCP configuration, and browser control, with hooks for state management and user interactions.

A standalone Agent Skill that guides design and implementation of distinctive, high-craft game UIs in Godot 4.7, combining art direction, interaction design, native implementation, and visual review without requiring external tools or dependencies. It includes 31 focused knowledge modules, reusable templates, two worked briefs, optional local validation scripts, and 12 evaluation cases to help produce original visual grammar rather than copying reference aesthetics.

gox-code-rules enforces team coding standards for Go (HTTP-gin framework, CLI, config, migrations, scaffolding), frontend (React/Vue/TypeScript), shell scripts, and skill authoring through 5 embedded skills that auto-activate during coding sessions. The plugin provides guidelines for architecture, naming conventions, SKILL.md layout, and general engineering best practices via a SessionStart hook.

This Claude Code plugin provides guided integration support for Halliday Payments, offering a `/halliday` command with flows for learning about crypto deposits and fiat-to-crypto onramps, cloning sample applications across multiple frameworks, validating integrations, and checking payment status. It bundles reference documentation, queries the live Halliday API for supported chains and assets, and runs an allowlisted wrapper script for API access.

Hit 'Em With The Docs is a self-managing documentation system that automatically organizes markdown into 15 specialized domains, applies metadata, checks for broken links, and reports documentation health scores. It integrates with Claude Code, GitHub Actions, and optionally pairs with semantic-pages for read-only AI search while maintaining single-writer control over the documentation tree.

This Claude Code plugin enforces house-style conventions for written prose and commit messages by flagging violations like em dashes, non-ASCII bytes in PowerShell, buzzwords, and filler phrases across files and git/gh commands. It ships with a default rule set in JSON format and merges optional user-defined rules to customize or override style checks.

This Claude Code plugin generates single-file HTML pages that explain agent outputs with plain language, meaningful visuals, and accessibility features designed for readers with ADHD or limited working memory. Each page follows a consistent structure with an up-front answer, visual elements per section, and optional decision-making components, all validated by an automated checker before delivery.

ido4shape guides product managers and founders through structured specification discovery via conversational probing of problems, assumptions, and stakeholder tensions, then synthesizes a formal strategic spec with capability decomposition, dependency mapping, and success conditions. It reads existing project materials, maintains a persistent knowledge canvas across sessions, and produces specifications designed to feed into AI coding agents for reliable delivery.

This plugin reformats Claude Code agent replies into a standardized, ASD-STE100-inspired structure with key-first bullets, one-sentence conclusions, and six numbered sections (Done, InProgress, Questions, Todos, Pending, Backlog) to make information scannable in under five seconds. It processes replies via session hooks and shell execution to enforce consistent, concise formatting across any language.

insight-blueprint is a Python MCP server for hypothesis-driven data analysis that manages analysis designs, data catalogs, and review workflows through Claude Code or any MCP-compatible client. It provides 10 analysis skills, 18 MCP tools, and a WebUI dashboard for browsing designs and domain knowledge across the full analysis lifecycle.

Three standalone agent skills for managing repository documentation, intent, and instruction files: `intent` interviews you into and audits an INTENT.md to track why the repo exists and catch code drift; `docs-rebuild` restructures monolithic markdown so agents can navigate it with verified links; `md-refactor` audits and refactors agent instruction files (AGENTS.md, CLAUDE.md, policy docs) for effectiveness with a ledger of changes.

invero-pm is a multi-agent project management system for construction and fit-out projects that maintains a live dashboard of drawings, schedules, costs and compliance state, then briefs specialist agents (cost, programme, tender, scope, health checks, code compliance, etc.) through a strict review and sign-off loop with statutory deadlines first. The system adapts to 60+ jurisdictions via extensible knowledge packs covering building codes, payment law, safety regimes and local terminology, with Australia in depth and the US as a starter pack.

IssueBadge agent skills teach Claude to issue, verify, and manage digital certificates and badges through an MCP server, with support for OAuth 2.1 authentication, PDF downloads, QR codes, and revocation workflows. The plugin includes 2 skills covering MCP and REST API integration paths, connecting to the hosted issuebadge-mcp.issuebadge.workers.dev server.

k12mcp provides structured guidance on K-12 compliance requirements including FERPA, COPPA, state privacy laws, and DPA readiness through 8 specialized skills and a multi-step compliance-review agent. The plugin connects Claude Code to a backend MCP server, offering PII scanning, SIS integration details, rostering protocols, and procurement pathway recommendations with a credit-based usage model.

Kanon reports every instruction file that governs a Claude Code session, where each one came from, and which expected files never loaded. It classifies sources as user-owned, project-based, or foreign (from dependencies), and tracks changes to files across sessions.

A Claude Code plugin that tracks lines written by the coding agent in a git-backed ledger, recording which ones you've explained in your own words. It uses a PostToolUse hook to monitor Write/Edit operations and provides a skill to guide you through explaining blocks of agent-written code, tracking your comprehension debt over time.

Lawgical provides Vietnamese legal research within Claude with precise citations down to individual articles and clauses, drawing from 150,000+ legal documents. It includes specialized skills for contract drafting, contract review, and compliance checklist creation.

A teaching system that combines the teach and visualize skills with researcher and diagram-maker subagents, delivering interactive lessons with LaTeX, Mermaid diagrams, and graded quizzes in a browser viewer synced to your terminal session. The plugin includes an MCP server that handles quiz grading, free-form questions, and diagram rendering to support a structured learning workflow.

An AI-powered tutor plugin that adapts Claude into a personal learning companion with structured pedagogy, spaced repetition, and persistent learner profiles. It offers five learning modes (system learning, concept explanation, quick review, quiz, study planning) with three adaptive teaching styles (progressive, Socratic, structured) and pre/post testing to track knowledge retention.

lecture-sync monitors a lecture materials folder for changes and automatically organizes content artifacts by subject using customizable formats (chapter supplements with quizzes or example-focused collections), generating knowledge graphs and Gmail notifications. It optionally incorporates lecture video transcripts into the organized materials.

Lesson Builder is a standalone Agent Skill that generates complete bilingual TEFL/ESL courses from a single natural-language prompt, producing lessons, exams, flashcards, conversations, pronunciation drills, worksheets, and syllabi through a research-grounded buddy workflow. It orchestrates discovery via Tavily Research and NotebookLM to ground course content in peer-reviewed pedagogy, then invokes Claude Code agents (spec, plan, tasks, implement) to produce standalone HTML courses with integrated AI avatar videos and images.

This plugin provides three optional hooks that run at session start, before certain tool uses, and after document edits to warn about sensitive files, destructive commands, and documentation quality issues. It integrates with common doc frameworks (MkDocs, Docusaurus, Sphinx, mdBook, Mintlify, Antora) and linters (Vale, markdownlint) to guide Claude's behavior during documentation work.

LibreCopy-hooks provides three session and tool-use hooks that enforce safety checks: a SessionStart summary of documentation tooling, a PreToolUse confirmation before touching secrets files or running destructive commands, and a PostToolUse check for empty writes, lint commands, and matching tests. All hooks read project files and tool JSON input to provide real-time feedback without logging or network calls.

Lightkeeper is a Claude Code plugin that maintains an Obsidian-compatible knowledge map of your codebase stored outside the repo, using hooks to ensure Claude reads the map before editing code and updates it before stopping. The plugin keeps the map honest through five behaviors: session primers, prompt hints, pre-edit pointers, Stop reminders, and maintenance signals that flag outdated or inconsistent notes.

This plugin assists Spanish civil law firms by managing case portfolios and drafting legal documents—including complaints, responses, appeals, and enforcement proceedings—while verifying citations against current jurisprudence and legislation through the Jurisprudenciator connector. It integrates with document storage, email, calendar, and e-signature services to support end-to-end case management compliant with LO 1/2025, the current LEC, and EGA standards.

A Spanish labor law litigation plugin for law firms that drafts procedural documents (conciliation forms, prior claims, lawsuits for dismissal, salary claims, disability, collective proceedings) and appeals under LRJS, ET, and Social Security law, powered by the included Jurisprudenciator connector for case law, legislation, collective agreements, and corporate registry access. The plugin manages case portfolios, learns the firm's writing style from sample documents, and integrates with Google Drive, Microsoft 365, Dropbox, Box, and DocuSign for document and email management.

A Claude Code plugin for Spanish criminal law practices that covers the complete litigation cycle from arrest and detention through trial, appeals, and execution, with 48 skills integrated with the Jurisprudenciator connector for live jurisprudence, legislation, and case citation verification. It enforces strict compliance with LECrim, CP, and recent reforms (LO 1/2025, LO 1/2026) and restricts all legal data sources to Jurisprudenciator to prevent citing errors in criminal proceedings.

Generates a navigable, version-controlled multi-page Markdown wiki from a codebase using a two-pass outline-then-fill approach across parallel agents, consuming a code-atlas index when available and providing on-demand retrieval as Claude's context memory. The plugin produces static Markdown with Mermaid diagrams, per-page source provenance, and git-blob hash-diff staleness detection, while a SessionStart hook automatically loads the wiki index into context at session start.

An LLM-maintained markdown wiki that enforces citations via footnotes, prevents edits to raw sources with a PreToolUse hook, and tracks disputes and freshness with mechanical linting. It implements Andrej Karpathy's LLM Wiki pattern with the failure modes closed.

Logbook creates a live browser page during Claude Code sessions that displays what awaits user input, what was built, what was tested, and what decisions were made—no model tokens spent on recording. When the session ends, the same page becomes a self-contained HTML report with full session history from transcripts and git commits.

Manar audits your website for SEO and AI-visibility compliance, then writes fixes directly into your codebase for Next.js, Astro, ASP.NET Core, and static sites via pull requests. It measures whether ChatGPT, Gemini, Perplexity, and Claude actually mention and cite your content, tracking results across releases.

md-linker automatically tracks cross-references between markdown documents and detects when linked content becomes stale, preventing Claude from reasoning from outdated information across sessions. It maintains a link graph, marks stale references in frontmatter, and uses a Sonnet sub-agent to resolve stale content before the main agent reads files.

mememo is an MCP server that indexes codebases with code-aware embeddings and semantic search, enabling Claude Code to store memories, search code semantically, build typed call/import graphs across 14 languages, and answer architectural questions with citations. It runs locally with private embeddings powered by torch and sentence-transformers, integrating via passive hooks to auto-capture and inject relevant context during development.

Memolok is a Decision Record platform that captures organizational decisions with their context, alternatives, and reasoning as immutable Memolok Decision Records (MDRs) sealed in a Decision Ledger. The Claude Code plugin provides 17 skills for recording, reviewing, committing, and tracking outcomes of decisions, plus an agent for querying ledgers without cluttering the conversation.

The memory-bank plugin maintains a structured directory of Markdown files that preserve project context across Claude Code sessions, allowing you to initialize and update documentation like project briefs, active context, and progress. It includes a stop hook that reminds you to update the memory bank if project files have changed since the last update.

memory-wiki transforms claude-memory's weekly rollups and flat concepts into a cross-linked, searchable wiki with a symptom index and component map injected at every session start, requiring no tool calls for recall. It reads memory outputs without modification, scaffolds the wiki structure, ingests rollups into typed pages with an append-only ledger, and provides a linting tool that audits links and schema without making changes.

Nagomi is a forensic-grade PubMed research engine accessible via slash commands in Claude Code, offering PICO-structured search, evidence synthesis, critical appraisal, citation verification, and reference export using a Python MCP server that queries NCBI directly. It provides study design classification, trustworthiness analysis with Oxford CEBM hierarchy, GRADE-style confidence ratings, and fuzzy-matching citation verification without requiring an API key.

NarrateAI DemoMaker records and narrates demo, tutorial, onboarding, and product-walkthrough videos of your running web app directly from Claude Code, with cinematic zoom, smooth cursor animation, and AI voiceover in one prompt. The plugin bundles an MCP server that drives a headless browser through your app and ships a fully narrated MP4 via the NarrateAI pipeline.

NerdyGeek is a version-aware MCP server and Claude Code plugin that helps coding agents fetch current official documentation, resolve version context from project files, and detect deprecated APIs. It provides three core tools—search_docs, diff_docs, and scan_deprecations—that return compressed, source-backed answers with confidence scores and cache tracking.

newsroom-core is a Claude Code plugin that provides investigation orchestration, evidence logging, timeline building, and data analysis tools for investigative journalists to manage complex stories before publication. It includes four agents and two skills that work together with shell access and post-tool-use hooks to streamline the investigative workflow.

This Agent Skill helps Dutch government developers prepare repositories for open source publication by generating essential files like publiccode.yml, README.md, CONTRIBUTING.md, SECURITY.md, and LICENSE based on project information. It also checks for the presence of required open source files and updates existing documentation files where necessary.

This Claude Code plugin provides an evidence-based health, fitness, nutrition, and longevity advisor agent tailored for adults 50+, using local Markdown context files to track health information and maintain session continuity. The agent enforces clinical rigor (noise-floor discipline, collection-cost transparency, safety boundaries) and scopes Bash access to permit only safe local analysis while blocking destructive and network commands.

This plugin orchestrates a complete research-to-publication pipeline, guiding users through ideation, experiment design, implementation, evaluation, and LaTeX manuscript generation in field-specific conventions. It enforces reproducibility by tracing all paper numbers to real experiments and iterates the draft through simulated peer reviewers until acceptance before handling camera-ready revisions.

A Claude Code plugin that routes document parsing across three backends—markitdown (free, MIT), Docling (tables and scanned PDFs), and LlamaParse (cloud)—with smart escalation based on document type and quality metrics. Includes an interpret tool that pipes parsed markdown into Claude for summarization and extraction tasks, plus chunking and diagnostics utilities.

Patristic provides semantic and lexical search over a Russian Orthodox patristic corpus of ~2,100 works across 86 authors, with tools for passage retrieval and citation generation from logospatrum.com. A subagent and skill enforce a disciplined two-step workflow: search returns candidates, then the main loop retrieves full passages to ensure accurate citations.

Plainspoken helps Claude write like a person by checking text against anti-AI patterns, enforcing plain language and voice consistency, verifying claims against sources, and tracking writing quality improvements through a kaizen loop. It includes four isolated judges, multiple specialized agents (cold-reader, claims-checker, delivery-checker), and a 232-case evaluation set to ensure text sounds authentic and honest.

This Claude Code plugin answers employee questions about AI-tool policies by retrieving relevant sections from a local folder of Markdown documents, determining which policy is in effect on a given date, and providing citations while blocking invented information. The plugin includes a grounding check that ensures all cited sections, links, and contacts actually exist in the source documents.

ProductForge researches an existing software product through public sources, reconstructs its features and workflows with explicit confidence levels, then generates an evidence-backed engineering blueprint including architecture decisions and a phase-by-phase implementation plan. The plugin traces every requirement back to its source, separates confirmed facts from inferences, and prepares a new Claude Code session to begin building from phase zero.

This bilingual plugin provides structured product review analysis for PRDs, UX designs, and workflows by identifying logic defects, interaction risks, and missing edge cases. The plugin uses an agent-based approach with local skill execution and shell utilities to produce audit reports with actionable insights.

Turns Claude into a personal programming professor that teaches concepts step-by-step, explains codebases to junior developers, quizzes understanding, and tracks learner progress across sessions. The plugin includes skills for lessons, onboarding, code walkthroughs, quizzes, warm-ups, and progress tracking, with automatic session reports and a persistent learner profile stored locally.

Project Pilot is a Claude Code plugin that maintains a persistent `.pilot/` knowledge base capturing your project's architecture, decisions, conventions, and current state across sessions. It tracks file edits and bash commands via hooks, auto-synthesizes changes into living documentation, and warns before editing critical paths you designate.

prose-mint scans files, pull request bodies, and piped text for structural patterns common in AI-generated writing—em dashes, arrows, not-X-but-Y phrasings, bold-colon openers, and AI-attribution boilerplate. It functions as a linter for prose style and composition, flagging these patterns to help maintain consistent writing conventions.

This plugin bundles Publish.fun's MCP server and agent skill to let Claude submit, revise, and track research papers at Publish.fun, an AI-native research journal that reviews submissions with frontier models and web search. It also lets Claude cite published papers and prepare manuscripts in Markdown, with safeguards to confirm submissions and revisions before sending.

rails-context creates per-directory AGENTS.md context files for Rails applications describing MVC layer conventions, with one-line CLAUDE.md imports for Claude Code and maintenance hooks to keep them in sync. It surveys a Rails app's structure, generates context files from actual code patterns, wires up drift detection via bin/audit_context, and supports upgrading from legacy symlink layouts.

This Claude Code plugin codifies a Rails app's UI into a design system documented under docs/design-system/, with per-section role tokens, component specs, page patterns, and drift auditing. It enforces the design system during Claude Code sessions via hooks, runs audit checks on view edits to catch token violations, and ships a design-reviewer agent for spec compliance.

RAINMOJO SO Lite is an SEO and AI Search Optimization (AISO) toolkit with 43 skills and 17 agents that audits whether AI platforms like ChatGPT, Perplexity, Claude, and Gemini can reach, read, and cite your website, while performing comprehensive SEO checks including keywords, technical audits, backlinks, and local SEO. It generates checklists and reports for WordPress drafts and works across Claude Code, Claude Desktop, ChatGPT, Gemini CLI, and other platforms.

RAINMOJO SO Lite is a comprehensive SEO and AISO (AI Search Optimization) toolkit with 43 skills and 17 agents that audits whether AI platforms like ChatGPT, Perplexity, Claude, and Gemini can reach, read, and cite your website, while also running classical SEO checks. It produces checklists and WordPress drafts without requiring API keys, relying only on Python for local execution.

Recap Studio turns a topic or coding session into a self-contained, dark-mode HTML explainer page that opens offline without requiring a server or internet connection. The plugin generates a five-minute read with a one-sentence answer, concept map, key ideas, glossary, and cited sources, plus a validation score for quality gates in CI.

repo-docs provides a framework for organizing repository instruction files through a hub-and-spoke model, keeping documentation correct and disclosed only when needed. It includes a skill and git commit hook that loads repository metadata from AGENTS.md and related files based on task context.

repo-docs is a portable agent skill that keeps repository instruction files (AGENTS.md, CLAUDE.md, docs/) correct and progressively disclosed, following the principle of putting information at the narrowest scope where it stays authoritative. It runs as a Claude Code or Codex plugin with a SessionStart hook that audits or maintains documentation structure automatically.

repo-docs organizes instruction files in a hub-and-spoke pattern, with a root CLAUDE.md that indexes documentation spokes and a drift hook that validates the structure on every commit. It bootstraps, adopts, or maintains the pattern with git integration and no external dependencies.

This collection of two standalone Agent Skills streamlines developer workflows: devto-post orchestrates end-to-end publishing of technical articles to dev.to with AI-generated covers and LinkedIn promotion, while repo-glowup audits and upgrades repository discoverability across READMEs, CI/CD, badges, and metadata. Both skills integrate with Claude Code and are installable individually via the plugin marketplace or as personal skills.

This plugin extends the Review Mode VS Code extension with AI agent skills and commands for collaborative annotation and revision tracking workflows in Markdown files. It integrates with Claude Code, GitHub Copilot, Cursor, and Codex through an MCP server for structured document review operations.

This Claude Code plugin guides structured exploration of complex problems by enforcing a Request for Comments (RFC) workflow with 8 sections including multiple proposed solutions, trade-offs, and open questions. It locks Claude into collaborative thinking mode with hooks that prevent code generation and force explicit multi-option analysis until a clean draft is ready for team review.

A comprehensive Claude Code plugin suite that transforms product owner requirements into versioned architecture documentation with decision records, risk analyses, and interactive 3D visualization, enforcing consistency through code-based gates and tests at every step. It delivers architecture as queryable source: ARCHITECTURE.md and supporting docs, ADRs/TDRs, risk analyses, developer handoff kits, Word briefings, and an MCP server (Alia) for IDE integration.

SDD is a Claude Code and Cursor plugin that enforces specification-driven development with RFC-2119 specs, requirements, and ADRs carrying stable identifiers and machine-checked traceability to code and tests. It provides a delivery pipeline with worker fan-out, review stages, and drift detection, operating purely on Markdown and JSON without external services.

sdlc-kit is a Claude Code plugin providing 11 skills that scaffold and maintain core software delivery artifacts—PRDs, ADRs, SDD specs, tasks, git worktrees, code reviews, and sprint retros—within an Obsidian-compatible vault. Each skill implements list/scaffold/transition contracts backed by SQLite indexing, auto-synced frontmatter validation, and a PostToolUse hook that cues re-sync after vault edits.

Series Continuity Director is an Agent Skill that tracks narrative state, character knowledge, camera perspective, and production decisions across scenes to maintain continuity in screen, comics, or prose works. It provides local Python tools to verify relationships, rebuild state, prepare submissions, and keep production records while letting the author and agent make all creative decisions.

session-continuity maintains cross-session memory for Claude Code projects through two committed Markdown files—SESSION_PRIMER.md for current state and LEARNINGS.md for discovered bugs—plus slash commands and hooks that surface the right knowledge at the right moment. This approach keeps sessions short and sharp by letting developers close work, start fresh, and get up to speed in two file reads instead of rebuilding context through re-prompting.

session-map creates a visual mind map of your project architecture within Claude Code, displaying every conversation and branch organized by the parts they work on, along with status, costs, and history. The plugin provides a browser-based interface that shows what's in progress, completed, waiting for you, and what API calls would cost, all accessible through simple commands and a welcome tour.

This plugin triages customer support questions, drafts contextual replies, escalates unresolved issues with full context, and converts resolved issues into help articles. It integrates with Slack, Intercom, HubSpot, Guru, Atlassian, and Notion to coordinate customer support workflows across multiple platforms.

Critiques designs, writes interface copy, reviews accessibility, synthesizes research and prepares developer handoff notes by connecting to Figma, Linear, Asana, Atlassian, Notion, Slack, and Intercom. The plugin includes seven skills for design critique, accessibility audits (WCAG 2.1 AA), UX copy writing, design system documentation, user research planning and synthesis, and developer handoff specifications.

Opens PDFs in an interactive viewer for reading, annotating, form-filling, and placing signatures, then downloads the marked-up copy. The SMILE-FX® PDF Viewer runs locally via a Node.js MCP server with no external account required.

This plugin helps product teams write specs, plan roadmaps and sprints, synthesize research insights, review metrics, and draft stakeholder communications. It integrates with 14 tools including Asana, Linear, Slack, Notion, Figma, Amplitude, and others to pull data and coordinate work across platforms.

A Chinese writing workflow plugin that enforces factual accuracy, removes AI voice patterns, and iteratively reviews content using separate sub-agents that cite evidence from the original text. It includes tools for AI similarity scoring, metadata/invisible character removal, and mechanical scanning of common writing issues.

Siyaq automatically indexes your project documentation and injects only the knowledge relevant to each prompt or file interaction, supporting multilingual matching (including Arabic) with BM25 relevance ranking and token budgets. It rebuilds its index automatically when docs change, tracks what helped and what's missing, and adapts injection depth based on context fill level.

This standalone Agent Skill provides two documentation reference skills: smelter-http-api-docs and smelter-ts-docs. The skills enable Claude to access structured API and TypeScript documentation during agent interactions.

This plugin enforces ASD-STE100 Simplified Technical English in Claude Code and Codex responses via a configurable strictness slider (10–100%), while preserving accuracy by marking and explaining any rule violations at higher levels. It intercepts prompts through a shell hook that injects formatting instructions, with rules stacking progressively across severity tiers and never sacrificing caveats or technical precision.

This Korean-language plugin detects and removes AI-generated text ("slop") from professional documents (meeting notes, reports, proposals, emails, PowerPoint/Word/Excel files) and narrative writing (reports, essays, blogs). It operates as a scanner and skill that integrates with Claude via a UserPromptSubmit hook to flag AI-generated content in real time.

This standalone Agent Skill provides two utilities: react-tsdoc generates TSDoc comments for React components and hooks, while readme generates StyleList94-style README.md files based on project type. Both skills are designed to automate documentation creation for JavaScript/TypeScript projects.

This standalone Agent Skill provides two utilities for documentation: react-tsdoc generates TSDoc comments for React components and hooks, and readme generates StyleList94-style README.md files based on project type. Both skills automate boilerplate documentation tasks without requiring external API keys or services.

Substrate is a multi-repo primitive registry that tracks reusable components (engines, APIs, pipelines, agents) across repositories via declarative manifests, compiling them into a dependency graph with staleness detection and orphan identification. It enforces outbound prose quality gates, memory consistency checks, and deliverable tracking at session start and on every file write to prevent parallel work and ensure new contributions plug into existing infrastructure.

This Claude Code plugin auto-generates a symbol index (glossary) of your codebase after /init or on demand, extracting classes, functions, and constants from Python, JavaScript, and TypeScript files. It hooks into PostToolUse to regenerate the GLOSSARY.md file automatically when your codebase changes, helping Claude understand your project structure while saving tokens.

The unpaged plugin renders Claude Code plans as interactive whiteboards on Unpaged, visualizing phases, tasks, and dependencies as connected canvases, then compiles an as-built record documenting what shipped versus planned and all implementation decisions. Comments mentioning @agent on the canvas are automatically pushed back to the session, and the listener checks every 30 seconds to keep plan and implementation in sync.

vault-administrator is a Claude Code plugin that helps Vault administrators manage live Vault instances through skills for configuration management, file staging, MDL scripting, and reporting. It integrates with the vault plugin's login system and optionally uses the Vault CLI for operations like pulling configuration, managing staged files, executing metadata changes, and analyzing reports.

This plugin automatically injects the nearest `vector.md` file into Claude's context whenever a file is read, ensuring design constraints and conventions are always visible to the model. It deduplicates injections, handles context compaction, and remains workspace-bounded while supporting agent-aware isolation of seen-sets.

vibe-ops is a Claude Code plugin that sets up and reconciles repositories and monorepos while authoring governance documentation (AGENTS.md, ADRs, RFCs, plans, tasks, licenses, and READMEs) from a shared source of truth. It includes 2 agents, 11 skills, 8 hooks, and 1 MCP server to manage repository organization and governance templates.

Adds the public Vocenya Docs MCP server and two skills to Claude Code for building on Vocenya, an AI receptionist for small businesses. The plugin includes endpoints for API documentation lookup, skills for REST API authentication and webhook management, and guidance for integrating Vocenya's website live chat across different frameworks.

Generates website compliance documents and cookie consent banners tailored to six jurisdictions (Israel, EU/EEA, UK, US federal, California, Canada) using reusable project profiles and AI-powered prompt templates. The toolkit includes a browser-based studio, a working cookie-consent banner in Hebrew/Arabic/English/Russian, and downloadable packages for WordPress or static sites with zero runtime dependencies.

This skill provides coherent web art direction and implementation guidance with 12 distinct visual styles and 18 techniques, routing tasks through brief analysis, style selection, and verification without enforcing any specific framework or library. It covers marketing pages, commerce, portfolios, applications, and data visualization while maintaining semantic integrity and accessibility.

Whalory is a copywriting methodology plugin that brings a writer's working method and brand voice profiling to any AI assistant, with local English and Persian linters that check copy on your machine without external dependencies. It includes two subagents (writer and blind editor), nine skills, an MCP server, and operates entirely with Python standard library using a SessionStart hook to initialize its copywriting workflow.

This plugin automatically lints content and locale files with Whalory after Claude writes or edits them, surfacing issues as context for follow-up fixes without blocking or modifying files. It supports Markdown, MDX, HTML, CSV, TSV, and multiple locale file formats while detecting Persian and English content line by line.

whats-new brings Claude Code feature discovery into your workflow by offering contextual recommendations, random daily suggestions, and proactive hints about new capabilities as you work. The plugin eliminates context-switching by surfacing relevant features at the moment you need them, powered by a knowledge base of 50+ curated entries updated daily via GitHub Actions.

work-order is a specification and reference implementation for a ticket-as-contract layer that standardizes how tickets carry problems, binding solutions, settled decisions, and scope boundaries. It provides a conformance validator, file binding for Markdown-based tickets, and optional extensions like sprint mechanisms and decision-list formatting.

This plugin enforces a consistent voice and style across drafts and outgoing messages through configurable rules applied at draft time and send time. It provides a multi-pass drafting pipeline for long-form content, send-time linting for Slack/email/docs, and learns correction patterns to surface recurring voice violations.

writing-register applies humanizer patterns and optional personal voice rules to steer Claude Code's writing style at three points: before generating output through system instructions, after writing via automatic rewrites of commit messages and PRs, and on-demand through the `wr humanize` command. The plugin uses hooks to intercept tool use and prompt events, integrating with git workflows while keeping all processing local.

UI UX Pro Max is a searchable local database of 79 UI styles, 192 color palettes, 74 font pairings, 25 chart patterns, and 22 implementation stacks (covering React, Vue, Svelte, Flutter, Tailwind, and more) designed to assist with page layouts, components, color schemes, typography, and accessibility. The plugin includes an AI-powered Design System Generator that analyzes project requirements and generates complete, tailored design systems with recommended patterns and styling guidance.

Understand Anything analyzes codebases and knowledge bases with a multi-agent pipeline to build an interactive knowledge graph showing files, functions, classes, and dependencies. The plugin provides a visual dashboard for exploring architecture, searching semantically, analyzing change impacts, and understanding business logic across projects of any size.

ruflo-adr manages the full lifecycle of Architecture Decision Records—creation, indexing, compliance verification, and relationship tracking—by storing ADRs as markdown in a hierarchical AgentDB with causal edges for supersedes/amends/depends-on relationships. It includes git-diff scanning for ADR compliance violations, semantic search, and reconciliation for deleted ADRs.

ruflo-docs automates documentation generation, drift detection, and API docs creation from TypeScript/JSDoc code using the Haiku model for cost efficiency. It drives a background `document` worker that generates full project docs, API references from JSDoc/TSDoc/OpenAPI, and detects when documentation falls out of sync with implementation.

The AAS Accessibility & Inclusive UX bundle packages 8 skills for Claude Code that help developers build more accessible and inclusive user experiences. The plugin provides guidance and tools for implementing best practices in accessibility standards and inclusive design.

This Claude Code plugin bundle provides 9 editorial skills for creating, managing, and presenting AAS (Agentic Awesome Skills) documents and presentations. It requires no external API keys or tools, operating entirely within Claude's native capabilities.

This bundle collects 10 skills for localizing content and supporting international growth strategies within Claude Code. It provides reusable components for language adaptation, market analysis, and global expansion workflows without requiring external API keys or services.

This Claude Code plugin bundles five privacy and compliance engineering skills from Agentic Awesome Skills to help developers implement privacy-preserving architectures and meet regulatory requirements. It provides tools for privacy assessment, compliance checking, and engineering practices without requiring external API keys or services.

This plugin bundles six design skills for Apple platform development, providing editorial guidance and best practices for creating native iOS, macOS, tvOS, and watchOS applications. It requires no external API keys or third-party services, making it immediately usable within Claude Code.

This Claude Code plugin bundles six editorial skills designed to assist with creative direction and content strategy. It integrates directly into Claude Code's workflow to support editorial planning and creative decision-making.

Creates editorial-quality diagrams including architecture, flowcharts, sequences, state machines, ER models, timelines, swimlanes, quadrants, radars, loops, trees, org charts, Venn diagrams, pyramids, Gantt charts, and Wardley maps, outputting self-contained HTML+SVG files with minimal styling and optional motion. The plugin converts existing draw.io, Mermaid, or Excalidraw diagrams to chosen formats and detail levels while maintaining semantic pattern separation from layout.

This Claude Code skill audits and rewrites prose that reads as machine-generated across markdown files like READMEs, changelogs, PR descriptions, and blog copy, offering detect-only, rewrite, and edit-in-place modes. It uses a pattern catalog covering formatting, sentence structure, and machine fingerprints, with tiered vocabulary tables and context/voice profiles, while explicitly avoiding false authorship classification and imposing house styles.

This plugin automates C4 architecture documentation by analyzing code from the bottom up to identify components, map containers, and generate context diagrams. It orchestrates multiple agents to synthesize architectural insights and produce comprehensive architecture diagrams and documentation.

This plugin generates documentation and explanations for codebases through automated doc generation and tutorial creation, with 2 commands and 3 agents to handle different documentation tasks. It requires no external APIs or tools and operates entirely within Claude's capabilities.

This Claude Code plugin automates documentation generation by creating OpenAPI specifications, Mermaid diagrams, tutorials, and API reference documentation through a set of coordinated agents and skills. It requires no external dependencies or API keys, making it fully self-contained for developers who need to quickly document their code and APIs.

This plugin provides HADS (Human-AI Document Standard), a semantic tagging convention that makes documentation readable for both humans and AI models, along with the grounded-vault pattern for knowledge storage with per-claim provenance links and git fingerprints. It enables zero-token drift detection by tracking documentation changes at the claim level.

This Claude Code plugin provides two agents for functional programming education and development using Elixir, OTP patterns, and the Phoenix framework. It covers distributed systems concepts and practices without requiring external APIs or tools.

This plugin provides HR policy documentation, legal compliance templates covering GDPR, SOC2, and HIPAA standards, employment contracts, and regulatory documentation through integrated agents and skills. It enables organizations to generate, manage, and ensure compliance with employment law and data protection requirements without external dependencies.

This plugin validates compliance with SOC2, HIPAA, and GDPR standards through automated scanning, secret detection, and regulatory checklists. It provides compliance validation and documentation generation without requiring external API keys or services.

This plugin provides three agents to handle SEO-focused content creation, planning, and auditing with E-E-A-T (Experience, Expertise, Authoritativeness, Trustworthiness) optimization. It operates entirely locally without external API dependencies or integrations.

This plugin analyzes and optimizes web pages for technical SEO, including meta tags, keywords, page structure, and featured snippet formatting. It contains four agents that assess and improve SEO elements to help pages rank better in search results.

A teaching skill that walks through setting up cryptographically signed audit trails on Claude Code tool calls using Ed25519 signatures, JCS canonicalization, and Cedar policies. Includes runnable examples, tamper detection walkthroughs, and CI/CD integration patterns as a precursor to the production-ready protect-mcp plugin.

A comprehensive UI/UX design plugin providing guidance on mobile and web application design, including design systems, accessibility compliance (WCAG 2.2), and modern interaction patterns across iOS, Android, React Native, and web frameworks. It offers design reviews, component creation, accessibility audits, and design system setup through multiple agents and skills.

This plugin provides an agent with three skills to audit pre-generated LP (limited partnership) statements before they are distributed, checking for accuracy and compliance issues. The auditing capabilities operate entirely within Claude Code with no external dependencies or third-party integrations.

The claude-md-management plugin provides tools to audit CLAUDE.md files against codebase state and capture session learnings into project memory. It includes a skill for periodic maintenance audits and a command for end-of-session documentation updates.

This is the Claude Code Plugins Directory documentation itself, explaining the structure, installation, and contribution guidelines for the plugin marketplace. It is not a functional plugin but rather the informational hub for discovering, installing, and developing Claude Code plugins.

Converts books and documentation (PDF, EPUB, DOCX, HTML, Markdown, RST, AsciiDoc, RTF, MOBI/AZW) into structured agent skills with a resident core framework index, on-demand chapter files, glossary, patterns, and decision cheatsheet. Ships four stdlib-only tools for extraction, validation, token budgeting, and plugin emission, plus an agent and commands to compile sources into reusable knowledge bases.

claude-coach is a personal Claude power-user coach that delivers a personalized cheat-code glossary on first use, then scans subsequent prompts to surface at most one high-impact tip when genuinely helpful. It includes a coaching decision tree, prompt rater, and three stdlib Python tools bundled as a Claude Code skill with a dedicated slash command and agent persona.

Analyzes frontend, backend, and fullstack codebases to automatically generate complete Product Requirements Documents, extracting routes, components, models, APIs, and data relationships. Uses Python scripts to scan code and scaffold structured PRD output with per-page documentation, API inventory, and enum dictionaries.

This Claude Code plugin generates CodeTour .tour files with persona-targeted step-by-step walkthroughs that link to specific files and line numbers, supporting 10 developer personas and all CodeTour step types. It includes built-in SMIG description formula support for structured documentation.

deepread analyzes supplied documents using five reading modes (quick, deep, map, feynman, book) and decomposes content into claims, reasons, and evidence with confidence labels. It generates evidence ledgers and knowledge maps to support comprehensive, evidence-first document comprehension without requiring external data sources.

Guides decision-making sessions for architectural plans by cross-referencing them against a project's documented terminology (CONTEXT.md) and architectural decisions (docs/adr/), then updates those files inline as decisions crystallise. Includes three Python linting and consistency-checking tools plus Matt Pocock's original interview discipline preserved verbatim under MIT.

Converts a research question into a strategically planned mini literature review delivered as a Word document, using free keyless PubMed and OpenAlex APIs with optional Consensus MCP enhancement. Output is an 8-section orientation guide designed as a launching pad for researchers entering an unfamiliar field, with configurable depth (5/10/20 queries) and built-in integrity checks for sequential execution and source tracking.

Converts long markdown files into single-file, self-contained interactive HTML with five specialized converters: long-form documents with sticky TOC and search, code reviews with 2-column diffs and annotations, and slide decks with presenter mode and keyboard navigation. A one-time design-system onboarding captures your brand palette and typography, ensuring consistent rendering across all conversions with WCAG-AA contrast enforcement.

A collection of 17 expert skills for HealthTech and MedTech regulatory affairs and quality management, covering ISO 13485 QMS, EU MDR 2017/745, FDA submissions, GDPR/DSGVO, ISO 27001, SOC 2, CAPA management, and audit compliance. The skills are organized across five strategic layers (leadership, core QM, risk/security, regulatory specialization, and audit/compliance) and work across Claude Code, Cursor, VS Code, and other AI agents without requiring external API keys.

research-summarizer is an agent skill for Claude Code, Codex, and Gemini CLI that summarizes academic papers, compares web articles, and extracts citations into structured research briefs. It processes research documents without requiring external APIs or tools.

This plugin provides a structured knowledge base extracted from "Spinning Up in Deep RL" by Joshua Achiam, covering RL fundamentals, model-free algorithm taxonomy, policy gradients, and six reference algorithms (VPG, TRPO, PPO, DDPG, TD3, SAC). Use it to apply Achiam's frameworks for debugging RL code, understanding algorithm derivations, and designing rigorous multi-seed experiments.

Generates a curated supplementary reading list from course syllabi by parsing topics and learning outcomes, then searching Consensus for recent peer-reviewed papers with audience-calibrated summaries and Bloom-aligned discussion questions. The skill structures topics into 6–12 sections with user checkpoints, runs targeted searches sequentially, and outputs a professionally formatted .docx with clickable paper links.

This Claude Code plugin helps users create new agent skills with proper structure, progressive disclosure, and bundled resources by combining Matt Pocock's original 3-phase workflow (Gather → Draft → Review) with stdlib Python validation tools and in-depth reference documentation. It surfaces skill authoring through a cs-skill-author persona agent and /cs:write-a-skill slash command that operationalizes Matt's review checklist to catch common mistakes.

This plugin provides 16 product management skills for creating PRDs, OKRs, roadmaps, user stories, and running sprints, retrospectives, and risk analysis through 11 commands. It operates entirely offline with no external API dependencies, offering frameworks for stakeholder mapping, prioritization, pre-mortems, and test scenario generation.

The pm-toolkit provides four core skills for product managers: resume review and tailoring against PM best practices, NDA drafting with jurisdiction-specific clauses, privacy policy generation covering compliance requirements, and grammar/flow checking with targeted fixes. It runs entirely locally without external API dependencies or third-party services.

A collection of 11 agent skills that help build great product interfaces by addressing typography, colors, layout, accessibility, UI polish, and product writing. Skills include reviews of existing interfaces, component variant generation, UI stress testing, and targeted improvements across design and accessibility domains.

This Claude Code plugin enhances instruction files by rewriting CLAUDE.md using <important if> blocks to improve how Claude follows your project guidelines. It's a single skill that helps clarify conditional instructions and boost instruction adherence.

The show-me plugin generates visual explanations of topics through concise diagrams, code-shape sketches, and focused HTML artifacts. It provides a single skill that helps clarify concepts by rendering them in diagram and code visualization formats.

Generates simple, visual HTML explanations of any topic using the /eli5 command, with large pictures and minimal text suitable for explaining concepts to a five-year-old. Requires no API keys or external tools, relying entirely on Claude's built-in capabilities.

This plugin generates a single interactive HTML page that organizes implementation plans as a hierarchical tree of claims, each supported by mockups, state machines, call stacks, schemas, or code snippets. Users can expand the tree progressively, answer embedded decisions, edit schemas inline, and respond with feedback that can be pasted back into Claude for refinement.

designlang extracts complete design systems from live websites—DTCG tokens, Tailwind/shadcn configs, Figma variables, motion tokens, and layout patterns—then grades consistency, generates brand-guidelines books, and measures visual fidelity against clones. Fourteen slash commands enable whole-site analysis, design-space positioning, head-to-head battles, theme recoloring, and shareable graded report cards with pixel-perfect diff scoring.

A deterministic documentation and code consistency auditor that inventories artifacts across README, CLAUDE.md, CHANGELOG, CI workflows, and package manifests, then reports drift via 9 structural checks (version strings, CI commands, references, cross-doc facts) resolved by human-adjudicated authority per fact class in sot-map.yaml. Read-only and advisory-safe — deterministic Critical findings can block releases, but LLM-judged findings are advisory-only and never fix anything.

Generates comprehensive API documentation from OpenAPI/Swagger specifications with interactive UIs and multiple export formats. Supports OpenAPI 3.0.3, Swagger UI integration, request/response examples, and automatic code sample generation.

The compliance-report-generator plugin provides a command and skill for generating compliance reports within Claude Code. It operates entirely with built-in capabilities and requires no external API keys or third-party services.

This Claude Code plugin generates comprehensive documentation for databases with best practices and multi-database support. It provides production-ready documentation templates and integrates via a single command for users with appropriate database access and permissions.

The database-schema-designer plugin helps developers design normalized database schemas by accepting application descriptions and generating complete CREATE statements, relationship mappings, and Mermaid ERD visualizations. It provides guidance on normalization levels (1NF through BCNF) and enforces industry-standard patterns including indexes and best practices.

FairDB Operations Manager provides comprehensive SOPs, incident response procedures, monitoring tools, and automation for managing PostgreSQL infrastructure. It includes three autonomous agents, six operational commands, and shell scripts for health monitoring, backup management, and compliance auditing.

Enhances file reading Skills by automatically generating production-ready code from requirements documents in multiple formats (PDF, Markdown, Word, Figma specs). Generates API endpoints with validation, tests, OpenAPI specs, and documentation for Node.js, Python, Go, TypeScript, and Rust projects.

The GDPR Compliance Scanner plugin scans applications and data systems to identify compliance gaps across data protection, privacy rights, consent management, and regulatory documentation requirements. It generates detailed reports with risk assessments, implementation guidance, and prioritized remediation recommendations.

This plugin automates contract review, risk analysis, and legal document generation through 5 parallel agents that perform clause extraction, risk scoring, compliance verification, and obligation mapping. It also provides skills for generating NDAs, privacy policies, terms of service, and conducting compliance audits across standards like GDPR, CCPA, and PCI-DSS.

Navigating GitHub is an adaptive skill that guides developers from GitHub setup and CLI installation to mastery through 9 hands-on lessons covering beginner, intermediate, and advanced git workflows. The skill assesses your experience level from your environment and tailors explanations and guidance accordingly, with built-in safety guardrails to prevent pushing to main or force-pushing without confirmation.

This Claude Code plugin automatically documents technical projects in Obsidian vaults by capturing progress, decisions, and insights as you work with Claude Code. It intelligently classifies projects by type (Hardware, Software, Woodworking, Music Synthesis), detects context from file patterns, and maintains cross-project relationships with scored wiki-links and optional Git integration.

This Claude Code plugin provides 24 operator-grade skills for safe clinical evidence review, adoption governance, and institutional rollout of OpenEvidence, with built-in safety boundaries ensuring clinical decisions remain with qualified professionals. Each skill is grounded in OpenEvidence's documented product, terms, and security policies, supporting workflows from onboarding and cost evaluation to incident response and production checklist verification.

This plugin provides four composable skills for governed editorial workflows: topic research with demand validation, editorial calendar balancing and scheduling, source-grounded article drafting with CMS adapters, and accessible SVG figure generation. All skills default to read-only research or local draft creation, with explicit owner approval required for scheduling, publishing, and credential use.

This plugin enhances Claude's web_search Skill to research cloud infrastructure best practices and automatically deploy infrastructure as code to multiple cloud providers. It generates Terraform/Pulumi configurations, applies them to GCP, AWS, Azure and other platforms, and sets up monitoring and documentation automatically.

This plugin generates comprehensive security audit reports with vulnerability assessments, compliance tracking across frameworks like OWASP/GDPR/HIPAA/PCI-DSS, and prioritized remediation roadmaps in multiple formats. It provides tailored reporting for different stakeholders including executive summaries, technical vulnerability details, security controls assessment, and trend analysis over time.

The security-incident-responder plugin guides users through structured incident response workflows including classification, containment, eradication, recovery, and post-incident analysis across incident types like data breaches, ransomware, and DDoS attacks. It provides response playbooks, evidence collection procedures, timeline construction, and remediation steps following established incident response frameworks.

Skill Creator is a Claude Code plugin that builds and validates production-grade agent skills using the 2026 AgentSkills.io specification and Anthropic best practices, featuring a 100-point marketplace grading rubric. It provides an interactive wizard for creating skills, automated validation against spec compliance, and auto-fix capabilities to improve marketplace scores.

This Claude Code plugin assists with SOC2 audit preparation by providing structured guidance and documentation support through a command interface. It helps teams prepare for SOC2 compliance by organizing audit requirements and relevant control evidence.

This plugin generates professional Statements of Work for AI automation projects with customizable templates covering scope, payment terms, and legal terms. It provides a single `/sow` command to create complete SOW documents tailored for AI project deliverables.

This plugin validates Claude Code plugin directory structure against the official Anthropic spec and Intent Solutions enterprise standard, checking plugin.json metadata, SKILL.md frontmatter and grading, commands and agents documentation, and file references. It provides comprehensive validation to ensure plugins conform to required schemas and conventions.