agent-memory
A four-tier memory ladder for Claude Code that accumulates recurrent facts from sessions into project and cross-project context, with strict promotion gates requiring multi-day/multi-project evidence and explicit human adoption before any changes commit. The system runs three hooks unattended (SessionStart, SessionStart, UserPromptSubmit) to inject context and capture raw transcripts, backed by five Python scripts that extract, promote, inspect, and validate claims against seventeen redaction patterns and open contradictions.
Install
Source: https://github.com/alirezarezvani/claude-skills/tree/HEAD/engineering/agent-memory
What it's made of
1 command · 1 agent · 1 skill · 3 hooks
- Commands
- 1
- Agents
- 1
- Skills
- 1
- MCP servers
- 0
- Hooks
- 3 · SessionEnd, SessionStart, UserPromptSubmit
What it needs & plugs into
- API keys
- none
- Paid services
- none detected
- External tools
python3- Talks to
- nothing external detected
Analyzed . 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 (alirezarezvani).