memory-curator
Skill4 starsChanged 21 days agoMITView on GitHub
Curate durable project memory by extracting verified decisions and discoveries while removing stale or sensitive details.
3 other projects use a version of this.
Adopted by
Most recently taken up by
- 3loveljw2/agent-identity-kittheir own version, first seen Aug 14, 2026
- wellorbetter/dsh-product-delivery-workflowtheir own version, first seen Aug 13, 2026
- ProjectWaja/claude-skillcrafttheir own version, first seen May 23, 2026
Other versions3 projects use a version of this.
Word for word · 0
None found.
Edited · 0
None found.
Wrote their own version · 3
- wellorbetter/dsh-product-delivery-workflow★ 1same name
- 3loveljw2/agent-identity-kit★ 0same name
- ProjectWaja/claude-skillcraft★ 0same name
How matches are found. Word for word: the exact same text. Edited: shares passages with this file (“99% same” means 99% of its five-word phrases also appear here; 10% or more counts). Their own version: similar meaning, or, weakest, the same name. We can’t tell which project had it first. Projects are counted among the 200 most-starred candidates, and mentions are links to the repository we found, up to 50 per site, so both counts are a floor.
In this repository45 other files it gives its agents.
- AGENTS.mdAGENTS.md
- CLAUDE.mdCLAUDE.md
- a11y-auditorSkill
- api-contract-reviewSkill
- backend-releaseSkill
- bug-diagnosisSkill
- client-copy-reviewSkill
- codebase-designSkill
- code-reviewSkill
- create-issueSkill
- create-prSkill
- create-taskSkill
- dependency-upgradeSkill
- deploy-evidenceSkill
- design-expertSkill
- design-system-docsSkill
- design-system-setupSkill
- domain-modelingSkill
- e2e-verifySkill
- evidence-auditSkill
- feature-discoverySkill
- frontend-releaseSkill
- generate-readmeSkill
- handoffSkill
- implement-taskSkill
- incident-responseSkill
- init-projectSkill
- live-docsSkill
- mcp-debuggingSkill
- plan-expertSkill
- planning-featuresSkill
- project-memory-initSkill
- project-memory-refreshSkill
- project-profileSkill
- project-releaseSkill
- proof-runnerSkill
- release-checklistSkill
- researchSkill
- security-hardeningSkill
- security-reviewSkill
- tddSkill
- wiki-forgeSkill
- wiki-initSkill
- wiki-querySkill
- wiki-syncSkill
The file
We split this skill into the rules it gives an agent
2 rules. About 0 are common, 2 are distinctive, and 0 are very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈2
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
- Distinctive: line 12
Keep durable decisions, non-obvious discoveries, conventions, and verified outcomes.
- Distinctive: line 14
Remove credentials, tokens, personal data, client secrets, and unnecessary raw transcripts.
≈ “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.
Files
Memory Curator
Intent: memory_curate — EN: curate project memory, summarize durable learnings. ES: organizar memoria del proyecto, resumir aprendizajes duraderos.
- Identify the memory store, ownership, retention rules, and whether the user explicitly authorized changes. If no write authorization exists, provide a proposed diff only.
For Git-backed project memory, proceed only when
.shipframe/memory/index.mdexists and local state fromproject-memory-state.sh status <repo-root>is notdeclined. A committed index signals shared opt-in for local curation on fresh clones. It never authorizes a push, branch, PR, or publication; those require the separateproject-memory-initpublication flow and fresh confirmation. - Keep durable decisions, non-obvious discoveries, conventions, and verified outcomes. Separate facts from hypotheses and mark time-sensitive facts with dates.
- Deduplicate and update the existing topic rather than creating conflicting entries. Preserve provenance and relevant file paths.
- Remove credentials, tokens, personal data, client secrets, and unnecessary raw transcripts. Never store secrets even when supplied.
- Validate links/paths and summarize additions, updates, deletions, and unresolved uncertainty.
- For project memory, keep one concise note per task under
.shipframe/memory/tasks/, update decisions only when a durable decision changes, preserve provenance/date, and never store transcripts or secrets. Engram is complementary and must never be synchronized with these files.
Host limits
Memory APIs differ by Claude Code, Codex CLI, and OpenCode installation. Use only an explicitly configured memory backend and its supported operations; never silently create or configure one.
Resumen (ES)
Guarda decisiones y hallazgos duraderos verificados; deduplica por tema, conserva procedencia y marca datos temporales. No escribas sin autorización ni almacenes credenciales, datos personales o transcripciones innecesarias.
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
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.