agentleFS
Sign inStart free
faizkhairifaizkhairi/claude-code-blueprint/memory/CLAUDE.md

claude-code-blueprint / memory

CLAUDE.md73 starsChanged 4 months agoMITView on GitHub

Markdown-based persistent memory system for AI assistants. Used by Claude Code to maintain context across sessions.

The file

We split this CLAUDE.md into the rules it gives an agent

7 rules. About 0 are common, 5 are distinctive, and 2 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈5

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 2

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  • could be automatic: a setting could enforce it
  1. Very specific:

    Session Start: Read core/session.md to recall where we left off

    line 16
  2. Distinctive:

    During Session: No need to constantly update files.

    line 17
  3. Distinctive:

    Session End: When the user says "save" or the session is wrapping up, update:

    line 18
  4. Very specific:

    Use templates/coding-template.md for new project entries

    line 26
  5. Distinctive:

    Don't commit sensitive data (API keys, passwords, PII, tokens)

    line 36
  6. Distinctive:

    Don't store code here, this is for context only

    line 37
  7. Distinctive:

    Don't delete diary entries (they form the session history)

    line 38

≈ “Common” is an estimate: at least 1 in 5 of the 26,980 CLAUDE.md files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
CLAUDE.md

Memory System

Markdown-based persistent memory system for AI assistants. Used by Claude Code to maintain context across sessions.

Purpose

This is NOT a code project, it is a structured knowledge base. Files here are read/written by Claude Code to maintain development context, user preferences, and project state.

Key Files

  • core/identity.md: AI personality and communication style config
  • core/preferences.md: User preferences, work patterns, tech stack knowledge
  • core/session.md: Working memory. Update at session end.
  • core/reminders.md: Persistent reminders that survive session changes
  • core/decisions.md: Append-only architectural decision log

Session Protocol

  1. Session Start: Read core/session.md to recall where we left off
  2. During Session: No need to constantly update files; focus on the work
  3. Session End: When the user says "save" or the session is wrapping up, update:
    • core/session.md with current topic, progress, next steps
    • core/preferences.md if new preferences or patterns were discovered
    • core/reminders.md: move completed items, add new ones
    • diary/ if this was a significant session (new feature shipped, major bug fixed, etc.)

Project System

  • Active projects tracked in projects/active/
  • Use templates/coding-template.md for new project entries
  • Max 10 active projects; oldest auto-archived

Integration with Claude Code

This repo complements (does not replace) Claude Code's native memory:

  • Claude Code's MEMORY.md stores technical facts (gotchas, patterns, configs)
  • This system stores relational context (preferences, session history, communication style)
  • Both are read at session start for full context restoration

Don't

  • Don't commit sensitive data (API keys, passwords, PII, tokens)
  • Don't store code here, this is for context only
  • Don't delete diary entries (they form the session history)

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

Check your setup, free