agentleFS
Sign inStart free
williamcorrea23williamcorrea23/sap-router-skill/.agents/skills/claude-mem/SKILL.md

claude-mem

Skill1 starsChanged 4 months agoMITView on GitHub

Persistent cross-session memory architecture for AI coding agents. Captures session activity, generates AI-compressed observations, maintains searchable timeline indexes, and provides progressive 3-layer context retrieval across restarts.

1 other project uses a version of this.

Adopted by

Most recently taken up by

The file

We split this skill into the rules it gives an agent

5 rules. About 0 are common, 5 are distinctive, and 0 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.

  • 0

    Very specific

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

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    When an interesting observation is identified from Layer 1, inspect its chronological neighbors to understand preceding context and …

    line 50
  2. Distinctive:

    Only fetch full markdown content for the specific observation IDs that are confirmed relevant:

    line 56
  3. Distinctive:

    When onboarding onto a new or unfamiliar repository, systematically prime the cognitive cache:

    line 65
  4. Distinctive:

    Read foundational architecture specifications and entry points.

    line 67
  5. Distinctive:

    When an agent session restarts, verify memory state:

    line 78

≈ “Common” is an estimate: at least 1 in 5 of the 350,495 skill 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
SKILL.md

Claude-Mem — Persistent Cross-Session Memory Architecture

Persistent context and cognitive cache across sessions for AI coding assistants (Claude Code, Gemini Antigravity, Cursor, Codex).

Overview

Claude-Mem captures session workflows, extracts key decisions, discoveries, and bugfixes into structured observations, indexes them chronologically, and provides progressive 3-layer retrieval to recover past context with minimum token overhead.

For full reference documentation, see:

  • Memory Search Guide
  • How Claude-Mem Works
  • Learn Codebase Workflow
  • Context Injection Rules

3-Layer Progressive Retrieval Workflow (MANDATORY)

Never fetch full observation details without filtering first. This ensures up to 10x token savings.

Layer 1: Search — Index with IDs

Query the memory index using targeted search terms and type filters:

search(query="<topic>", limit=20, project="<project-name>")

Output: Lightweight index table (~50-100 tokens per result) containing ID, timestamp, type, and title:

IDTimeTTitleTokens
#104210:15 AM🟣Configured ADT REST endpoints~65
#108802:30 PM🔴Fixed CSRF token expiration~50

Supported observation types (obs_type):

  • bugfix: Defect resolution and root cause
  • feature: Newly created capabilities and tools
  • decision: Architectural decisions and design rationales
  • discovery: Environment findings, ports, endpoints, credentials
  • change: Refactorings, file moves, and dependency upgrades

Layer 2: Timeline — Local Context Around Anchor

When an interesting observation is identified from Layer 1, inspect its chronological neighbors to understand preceding context and subsequent effects:

timeline(anchor=1088, depth_before=3, depth_after=3, project="<project-name>")

Layer 3: Fetch — Read Selected Observations

Only fetch full markdown content for the specific observation IDs that are confirmed relevant:

get_observations(ids=[1088, 1042])

Cognitive Priming (learn-codebase)

When onboarding onto a new or unfamiliar repository, systematically prime the cognitive cache:

  1. Scan directory structure and package manifests.
  2. Read foundational architecture specifications and entry points.
  3. Record high-level structural notes into the persistent memory store.
  4. Subsequent sessions query the cache via search instead of re-reading raw files.

Session Continuity & Memory Gating

In this repository, Claude-Mem integrates with scripts/memory_manager.py and MEMORY.md:

  • Active session context is maintained under 20 blocks / 100 lines total.
  • Archived decisions and milestones are indexed into long-term storage.
  • When an agent session restarts, verify memory state:
    python scripts/memory_manager.py verify --input MEMORY.md

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