Skip to content
Get packs

Compound Learning Documenter

Records one hard-won, non-obvious solved problem as a durable repo learning for future agents

Source Download skill
Repository
EveryInc/compound-engineering-plugin
Stars
25,850
Licence
MIT
Files
15 files

What it does

After a developer has solved a tricky problem, this skill makes the agent decide whether the reasoning behind the fix would be lost if only the final code and tests remained. If it would, the agent writes a short, grounded learning document under a solutions folder in the repository so the next person or agent can find it. If the lesson is already obvious from the code or existing docs, the agent writes nothing and explains why.

Great for

  • Developers who want to keep non-obvious lessons from debugging sessions in their repo
  • Teams that repeatedly rediscover the same pitfalls and want a searchable record
  • Maintainers of projects using agent workflows who want knowledge captured between sessions

Skip it if

  • Routine fixes or typo corrections, because the code and tests already explain them and the skill is designed to skip these
  • Repositories where the agent should not write files without review, since the interactive flow can add lines to project instruction files and rule files after consent
  • Sensitive projects where session content might contain secrets, since the documents are written into the repo and may be committed

Try asking

  • “/ce-compound the race condition in the job queue came from retries reusing a stale lock; the fix is not obvious from the code”
  • “/ce-compound mode:non-interactive depth:lightweight we chose to pin the parser version because newer releases change error codes”
  • “Document the reason we avoid the cache layer for billing data, now that the fix is verified”
Step-by-step guide: how to use Compound Learning DocumenterInstall, first run, examples, tips and FAQs

Install in 3 steps

  1. 1

    Download the skill

    The ZIP contains the 15 original files, the MIT licence and an attribution note.

    ce-compound.zip
  2. 2

    Add it to your agent

    Claude Code (all projects)

    mkdir -p ~/.claude/skills && unzip ce-compound.zip -d ~/.claude/skills/

    Claude Code / Agent SDK (this project only)

    mkdir -p .claude/skills && unzip ce-compound.zip -d .claude/skills/

    Claude apps (web/desktop): Settings → Capabilities → Skills → Upload skill, and choose the ZIP.

  3. 3

    Ask for the task

    No special command needed — the agent reads the skill's description and loads it when your request matches. Try one of the prompts above.

Get better results

  • Invoke it right after a problem is verified, and describe the context briefly so the agent grounds the document in the right session details
  • Run one learning per invocation; for a session with several lessons, run the skill once for each
  • Review the written document before committing, especially any instruction-file or pack changes offered in interactive mode

What's inside

  • assets/resolution-template.md
  • references/agents/best-practices-researcher.md
  • references/agents/data-integrity-guardian.md
  • references/agents/framework-docs-researcher.md
  • references/agents/pattern-recognition-specialist.md
  • references/agents/performance-oracle.md
  • references/agents/security-sentinel.md
  • references/agents/session-historian.md
  • references/assembly.md
  • references/concepts-vocabulary.md
  • references/enhancement.md
  • references/form.md
  • references/grounding-validation.md
  • references/lightweight.md
  • SKILL.md

Redistributed unmodified from EveryInc/compound-engineering-plugin under the MIT licence. The assessment was drafted with AI from the skill's files on 2026-10-29. Skills can instruct your agent to run code — read SKILL.md and any scripts before installing.