Skip to main content
skill CX Codex

extract-learnings — Claude Code skill: memory and installation

extract-learnings is a Claude Code skill that saves project knowledge and keeps memory notes organized. Record a finding or consolidate existing notes.

no license ★ 276 updated 2026-10-29

Install into Claude Code · Plugin marketplace All methods and platforms ↓
/plugin marketplace add gupsammy/Claudest
/plugin install claude-memory@Claudest
Open on GitHub

What extract-learnings does

For a one-off note, the skill checks existing memory for duplicates and chooses a suitable layer. During consolidation, it gathers context from memory files, CLAUDE.md and Git history, then runs a memory auditor alongside an agent that looks for new signals. If MEMORY.md does not exist yet, the skill creates it and skips the audit agent.

The memory hierarchy separates general preferences, project rules, short working notes and detailed topic files. MEMORY.md loads in every session; supporting details can be opened when needed. When a section grows beyond its limit, its pointers move to a cluster file. L2 and L3 notes live under the user's ~/.claude directory, so they are not shared with teammates. Project-wide rules belong in <repo>/CLAUDE.md.

The skill reads project memory and runs git log --oneline -20. Consolidation uses the claude-memory:memory-auditor and claude-memory:signal-discoverer subagents; with a new, empty memory, only the signal-finding agent runs. That distinction matters if you expect an audit of existing notes.

What it helps with

  • save project findings for later sessions
  • find duplicate or outdated memory notes
  • sort knowledge into the project's memory layers
  • move detailed references out of the always-loaded index

Example prompts for the agent

  • Remember this project’s build order for future sessions and save it in the right memory layer.
  • Consolidate project memory and identify notes that should be merged or retired.
  • Review recent project changes and record recurring findings for later sessions.

How to install extract-learnings

1Plugin marketplace

/plugin marketplace add gupsammy/Claudest
/plugin install claude-memory@Claudest

This installs the whole claude-memory plugin with all its skills

2Skills CLI (Vercel)

npx skills add gupsammy/Claudest --skill extract-learnings -a claude-code

3Manually: put the files in place

~/.claude/skills/extract-learnings/SKILL.md

Copy the whole skill folder; for one project use .claude/skills/extract-learnings/

Step by step, with commands for each platform: How to install skills in Claude Code: five ways

Questions about extract-learnings

How do I install extract-learnings in Claude Code?

Add the marketplace with /plugin marketplace add gupsammy/Claudest, then install the plugin with /plugin install claude-memory@Claudest.

What does extract-learnings do?

It stores project knowledge in a layered memory system and can review existing notes while looking for new findings. Notes range from general rules to detailed topic references.

Can teammates access all the saved memory?

Project rules in <repo>/CLAUDE.md are available to teammates. L2 and L3 notes live in the user's ~/.claude directory, so they remain personal.

Similar skills

All skills for Agent setup: memory, context, orchestration →

From the same set47

The whole set →
improve-skill improve-skill audits skill effectiveness, simulates user requests, and pinpoints missing capabilities in instructions. CX Codex +1 ★ 276repo stars performance-auditor The performance-auditor subagent inspects code for algorithmic, data-access, and resource bottlenecks. Use it after changes on hot paths or when you need to reason about how code scales. CX Codex ★ 276repo stars test-engineer test-engineer generates tests for critical business logic right after a feature or bug fix, skipping tests written purely to inflate coverage numbers. CX Codex ★ 276repo stars claude-memory-architecture claude-memory-architecture defines the technical architecture for the claude-memory plugin, detailing hook lifecycles, SQLite database operations, lock handling, and local vector indexing. CX Codex +1 ★ 276repo stars code-auditor code-auditor goes over the code you just changed, flags logic bugs, vague names and leftovers, then says in plain words what needs fixing. CX Codex ★ 276repo stars create-skill create-skill guides Claude Code through building a skill or slash command by clarifying its purpose and activation scenarios. Use it to draft a new instruction file. CX Codex +1 ★ 276repo stars security-auditor security-auditor finds exploitable vulnerabilities in auth, untrusted-input handling, secrets, and dependencies. Triggered after changes to auth, API endpoints, or file operations. CX Codex ★ 276repo stars create-agent create-agent generates configured subagent markdown files for Claude Code with frontmatter and system prompts. CX Codex +1 ★ 276repo stars

Project sponsors

Companies that keep this analytics open