re0-memo — Claude Code skill: lessons from a work cycle
Turns a finished, failed or disappointing work cycle into lessons, anti-patterns and quality gates the next pass can actually use.
npx skills add LilMGenius/paperthin --skill re0-memo -a claude-codeWhat re0-memo does
The skill fires when the artifact is done, broken or underwhelming and the useful output is what the next pass must learn. It reads the original objective, the final artifact, QA evidence, user complaints and local planning notes, then separates working assets from misleading progress: contracts, schemas, tests and services against panels, scaffolds and abstractions that only looked productive.
Then it generalizes. When feedback arrives as a pile of specific complaints, they get clustered upward into the shared pattern first, and the root cause is what gets recorded as the lesson — the specifics sit under it as evidence. Every failure is named as an anti-pattern, repeated ones become hard gates for the next cycle, and vague user direction turns into architecture vocabulary a fresh agent can read.
Output is local docs: a re0-memo with the lessons and a plan holding next-cycle contracts, gates and vocabulary. The skill explicitly forbids defending the artifact or writing a changelog instead of lessons. The final check is blunt: a from-scratch agent should tell what to preserve, what to discard and what to test first without reading the whole old session. Provenance stays local, so shipped artifacts don't narrate their scars.
What it helps with
- pull durable lessons out of a failed pass instead of defending the artifact
- collapse many user complaints into one root pattern
- turn repeated failures into gates for the next cycle
- sort reusable contracts and tests from scaffolds that only looked like work
- translate vague client direction into architecture vocabulary
Example prompts for the agent
- Run re0-memo on this cycle: what to keep, what to drop, which gates to set for the next pass
- Cluster the demo complaints into one pattern and write the anti-pattern with its gate
- Refresh the re0-memo and the plan for the next cycle from the QA results
How to install re0-memo
1Skills CLI (Vercel)
npx skills add LilMGenius/paperthin --skill re0-memo -a claude-code2Manually: put the files in place
~/.claude/skills/re0-memo/SKILL.mdCopy the whole skill folder; for one project use .claude/skills/re0-memo/
1Skills CLI (Vercel)
npx skills add LilMGenius/paperthin --skill re0-memo -a codex2Manually: put the files in place
.agents/skills/re0-memo/SKILL.mdFor all projects use ~/.agents/skills/
1Skills CLI (Vercel)
npx skills add LilMGenius/paperthin --skill re0-memo -a cursor2Manually: put the files in place
.cursor/skills/re0-memo/SKILL.mdCursor also reads .agents/skills/ and .claude/skills/
1GitHub CLI
gh skill install LilMGenius/paperthin re0-memo2Manually: put the files in place
.github/skills/re0-memo/SKILL.md1Skills CLI (Vercel)
npx skills add LilMGenius/paperthin --skill re0-memo -a gemini-cli2Platform CLI
gemini skills install https://github.com/LilMGenius/paperthin --path skills/coil/re0-memoManually: put the files in place
.devin/skills/re0-memo/SKILL.mdThe legacy .windsurf/skills/ path still works
Skills CLI (Vercel)
npx skills add LilMGenius/paperthin --skill re0-memo -a opencodeClaude.ai
Zip the folder and upload it in Settings → Capabilities → Skills
Step by step, with commands for each platform: How to install skills in Claude Code: five ways
Questions about re0-memo
How do I install re0-memo in Claude?
Run npx skills add LilMGenius/paperthin --skill re0-memo -a claude-code. The skill lands at ~/.claude/skills/re0-memo/SKILL.md.
What does re0-memo do?
It converts a finished or failed work cycle into lessons, anti-patterns, quality gates and vocabulary for the next pass. The output is local docs a fresh agent can act on.
How is re0-memo different from a changelog?
The skill bans changelogs outright — file lists and effort summaries are not lessons. Every lesson has to trace back to observed cycle evidence such as the objective, QA output, diffs or user feedback.
Source text
SKILL.md · MIT
--- name: re0-memo description: "Turn a finished, failed, or disappointing work cycle into portable lessons, anti-patterns, quality gates, and next-cycle vocabulary. Use after a build, QA pass, demo, user complaint, or abandoned attempt when the useful output is what the next pass must learn rather than the code itself." --- Turn a completed cycle into lessons the next cycle can actually use. ## Goal `re0-memo` extracts durable learning from a work cycle without defending the artifact that produced it. A cycle can run, pass tests, and still be the wrong thing. The output is not a changelog or therapy note: it is a local, evidence-backed record of what worked, what misled the build, and what gate the next pass must clear. Use `re0-memo` when the artifact is done, failed, disappointing, or ambiguous enough that the next agent needs the cycle's lessons more than its momentum. ## Workflow 1. Read the original objective, final artifact, QA evidence, user complaints, and any local planning notes. 2. Separate working assets from misleading progress: contracts, schemas, tests, services, vocabulary, and examples that earned reuse vs. UI, panels, scaffolds, or abstractions that only looked productive. 3. Generalize before recording: when feedback arrives as many specific complaints, cluster them upward into the shared pattern before writing anything down — many specifics against many items usually share one root gap. Record the root and its pattern family as the lesson; keep the specifics under it as evidence, not as lessons themselves. 4. Name each failure as an anti-pattern, not a mood. 5. Convert repeated or high-impact failures into quality gates for the next pass. 6. Convert vague user direction into architecture vocabulary a fresh agent can use. 7. Write or refresh local docs for the cycle: a re0-memo for lessons and a plan for next-cycle contracts, gates, and vocabulary. 8. Verify that a from-scratch agent could avoid the same failure from those docs alone. ## Rules - Do not defend the artifact. If it missed the product, say what missed. - Do not write a changelog. File lists and effort summaries are not lessons. - Do not transcribe a specific complaint as a lesson. A lesson that only fixes the reported case keeps the system human-in-the-loop: the same class of failure returns anywhere a person didn't point. Escalate every specific to the pattern that catches its whole class — the reporter's exact words are evidence, not the gate. - Preserve negative corpus. Failed paths are training data. - Prefer hard gates over advice. - Cite evidence from the cycle: objective, file facts, QA output, screenshots, transcripts, diffs, or user feedback. - If the next agent cannot act on it, it is not a lesson yet. - Keep provenance local; shipped artifacts should not narrate their scars. ## Verification Before finishing: 1. Every lesson traces to observed cycle evidence. 2. Every anti-pattern names a concrete failure mode and the gate that catches it. 3. No lesson is a single reported case in disguise: each covers a class, with the specifics that raised it sitting under it as evidence. 4. A fresh agent can tell what to preserve, what to discard, and what to test first without reading the whole old session.
© LilMGenius · MIT · LilMGenius/paperthin