sycophancy-correction
A standalone Agent Skill that detects and corrects sycophantic patterns (overly agreeable, ungrounded, or structurally flattering content) in LLM completions, prompts, and pipeline configurations through pattern classification and AI-powered rewriting. Distributed as an MCP server binary with hooks for extensibility, it exposes tools for detection-only scoring, combined detect-and-correct workflows, and specialized analysis of PMPO Reflect phase outputs.
Install
Installs via the community skills CLI (works with Claude Code and other agents).
Source: https://github.com/Know-Me-Tools/sycophancy-correction-skill
What it's made of
1 skill · 1 MCP server
- Commands
- 0
- Agents
- 0
- Skills
- 1
- MCP servers
- 1
- Hooks
- 0
What it needs & plugs into
- API keys
ANTHROPIC_API_KEY- Paid services
- Anthropic
- External tools
sycophancy-correction- Talks to
- nothing external detected
Facts extracted from the plugin's files. Prose generated by claude-haiku-4-5-20251001.
Is this your plugin?
Claim it to keep the card accurate and enter the weekly contest. Requires signing in as the GitHub owner (Know-Me-Tools).