multi-repo-learning
Skill1 starsChanged 5 months agoMITView on GitHub
Export, exchange, and import portable coding lessons across repositories without requiring a hosted service or daemon. Enable teams to share sanitized, portable lessons across repositories so prevention rules, fix patterns, and root-cause insights discovered in one project are safely reusable in another. This skill provides the workflows and tools for exporting local lessons, importing cross-repo exchange batches, and maintaining privacy throughout.
In this repository117 other files it gives its agents.
- AGENTS.mdAGENTS.md
- CLAUDE.mdCLAUDE.md
- advanced-orchestrationSkill
- architect-agentSkill
- implementer-agentSkill
- quality-councilSkill
- reviewer-agentSkill
- tester-agentSkill
- writer-critic-pairSkill
- authoring-skillsSkill
- api-endpoint-qualitySkill
- async-job-qualitySkill
- auth-qualitySkill
- db-migration-qualitySkill
- frontend-state-qualitySkill
- acceptance-criteriaSkill
- adaptive-flowSkill
- adaptive-prompt-selectionSkill
- adversarial-code-reviewSkill
- architecture-decision-recordsSkill
- brainstormingSkill
- brownfield-spec-enhancementSkill
- bug-fix-lifecycleSkill
- checkpoint-validationSkill
- cicd-integrationSkill
- clarify-before-codeSkill
- claude-code-hooks-packSkill
- code-context-packSkill
- code-intelligence-reviewSkill
- context-policySkill
- context-rich-implementationSkill
- crash-proof-planningSkill
- critique-pass-protocolSkill
- dependency-aware-task-orderingSkill
- deprecation-migrationSkill
- disciplined-diagnosisSkill
- docs-authorSkill
- executing-plansSkill
- external-skillSkill
- finishing-a-development-branchSkill
- git-guardrailsSkill
- goal-driven-executionSkill
- grill-user-before-buildingSkill
- guard-bypass-protocolSkill
- improve-codebase-architectureSkill
- incremental-reviewSkill
- issue-slicingSkill
- lessons-learned-dbSkill
- loop-engineeringSkill
- model-aware-configSkill
- model-weakness-memorySkill
- orchestration-workflowsSkill
- plan-driven-executionSkill
- plan-from-specSkill
- prd-from-contextSkill
- project-constitutionSkill
- prompt-architectureSkill
- prototype-before-commitmentSkill
- quality-engineSkill
- quality-execution-contractSkill
- The first 60 shown.
The file
Files
Multi-Repo-Learning
Export, exchange, and import portable coding lessons across repositories without requiring a hosted service or daemon.
Purpose
Enable teams to share sanitized, portable lessons across repositories so prevention rules, fix patterns, and root-cause insights discovered in one project are safely reusable in another. This skill provides the workflows and tools for exporting local lessons, importing cross-repo exchange batches, and maintaining privacy throughout.
When to use
- You discovered a valuable pattern or fix in one project and want to reuse it in another.
- A teammate maintains a related repo and could benefit from your team's lessons learned.
- You want to audit your local lesson DB for quality, expiry, or sensitive content before sharing.
- You received a lesson exchange batch from another repo and need to review and import it safely.
Inputs
- Source lesson entry or lesson database, commonly based on
templates/lesson-entry-template.md. - Target repository context: stack, domain, runtime constraints, and relevant quality packs.
- Privacy level for sharing:
public,internal,private, orredacted. - Optional exchange file matching
schemas/lesson-exchange-schema.json. - Optional golden example candidate using
templates/golden-example-entry.md.
Workflow
- Check quality — run
npm run lesson:checkornode scripts/lesson-exchange.mjs --export --dry-runto validate export readiness. - Redact before sharing — remove secrets, customer data, proprietary endpoints, raw logs, and repo-specific names unless they are intentionally public.
- Export — run
npm run lesson:exportornode scripts/lesson-exchange.mjs --export --output docs/lessons/exchange-export.json. - Review exchange batch — inspect
lesson_id,source_repo,area,severity,root_cause,fix_pattern,prevention_rule,tags, andprivacy_level. - Import safely — run
npm run lesson:import -- --input <file>ornode scripts/lesson-exchange.mjs --import --input <file>. - Promote when useful — convert high-signal imported lessons into golden examples using
templates/golden-example-entry.md. - Inject prevention — reference imported lessons in Smart Adapt / Model-Aware Config before future tasks.
Outputs
- Portable exchange JSON file validated by
schemas/lesson-exchange-schema.json. - Imported lesson markdown at
docs/lessons/imported-lessons.md. - Optional golden example entry for future prompt injection.
- Updated prevention rule or quality checklist for the receiving repo.
Failure modes
- Sensitive data leakage: exchange file contains raw logs, tokens, customer names, or private URLs. Stop and redact before export/import.
- Over-specific lesson: fix pattern only applies to one repo's internals. Generalize the root cause and prevention rule.
- Schema mismatch: imported JSON does not match
schemas/lesson-exchange-schema.json. Fix the batch before use. - Duplicate lesson: receiving repo already has the same prevention rule. Merge or cross-link instead of duplicating.
- Low-signal lesson: lesson lacks a concrete root cause, fix pattern, or prevention rule. Keep it local until clarified.
Verification checklist
- Exchange JSON validates against
schemas/lesson-exchange-schema.json. - No secrets, raw logs, customer data, private URLs, or real credentials are present.
- Lesson includes root cause, fix pattern, prevention rule, and tags.
- Receiving repo context was checked before promotion.
- Imported lesson links to relevant quality pack, Smart Adapt, or Model-Aware Config workflow.
- Golden example promotion includes before/after pattern and a prevention prompt.
Related assets
- Command:
commands/vibe-lesson-exchange.md - Schema:
schemas/lesson-exchange-schema.json - Sample:
templates/lesson-exchange-sample.json - Guide:
docs/multi-repo-learning.md - Examples:
examples/multi-repo-learning/README.md
Notes
Multi-repo learning is markdown-first and repository-local. It does not require a hosted service, daemon, telemetry upload, or runtime expansion.
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.

