proj-memory
Skill0 starsChanged 10 months agoMITView on GitHub
Set up and maintain structured project memory systems that track bugs with solutions, architectural decisions, key project facts, and work history. Use when setting up new projects, documenting decisions, tracking bugs and solutions, or maintaining project knowledge across sessions.
In this repository33 other files it gives its agents.
- architectSkill
- brainstormingSkill
- concise-planningSkill
- context-engineerSkill
- data-deprecation-analysisSkill
- data-modelerSkill
- data-pipeline-engineerSkill
- data-quality-frameworksSkill
- dbt-transformation-patternsSkill
- github-workflow-automationSkill
- guardianSkill
- implementerSkill
- json-to-pydanticSkill
- llm-app-patternsSkill
- navigatorSkill
- ops-managerSkill
- orchestratorSkill
- principal-data-engineerSkill
- product-manager-toolkitSkill
- prompt-librarySkill
- qa-testerSkill
- researcherSkill
- senior-data-engineerSkill
- software-architectureSkill
- spark-optimizationSkill
- sql-optimization-patternsSkill
- state-managerSkill
- subagent-driven-developmentSkill
- thinkerSkill
- top-web-vulnerabilitiesSkill
- verification-before-completionSkill
- web-design-guidelinesSkill
- workflow-managerSkill
The file
Show its rules and text
We split this skill into the rules it gives an agent
13 rules. About 0 are common, 11 are distinctive, and 2 are very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈11
Distinctive · estimate
This author's own choices. Worth a look.
2
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Very specific: line 36
Memory Infrastructure: - Create docs/project_notes/ structure
- Distinctive: line 37
Bugs Management: - Document bugs, root causes, solutions, prevention
- Distinctive: line 98
Create 4 memory files (bugs.md, decisions.md, key_facts.md, tracks.md)
- Distinctive: line 99
Add templates from reference guide
- Distinctive: line 101
Start documenting as you work
- Very specific: line 183
Add memory section to AGENT.md:
- Distinctive: line 215
Search bugs.md for similar issues
- Distinctive: line 216
If found, apply known solution
- Distinctive: line 217
If new, investigate and document
- Distinctive: line 223
Check decisions.md for related decisions
- Distinctive: line 225
Document in ADR format
- Distinctive: line 232
If not found, document
- Distinctive: line 244
key_facts.md: - Verify URLs, ports, facts still current
≈ “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
Project Memory
Maintain institutional knowledge across your project lifecycle and team members.
When to Use This Skill
Use when:
- Setting up project memory system for first time
- Documenting architectural decisions and rationale
- Tracking bugs, solutions, and prevention measures
- Maintaining key project facts and configuration
- Logging completed work and milestones
- Configuring AGENT.md with memory protocols
- Quarterly review and refresh of project knowledge
Core Capabilities
- Memory Infrastructure - Create docs/project_notes/ structure
- Bugs Management - Document bugs, root causes, solutions, prevention
- Decision Records - Architectural Decision Records (ADRs) with context
- Key Facts - Configuration, URLs, ports, credentials pointers
- Work Tracking - Log completed tickets and milestones
- Agent Integration - Configure AGENT.md for memory-aware behavior
- Knowledge Maintenance - Quarterly refresh and pruning
Memory System Structure
Directory Layout
docs/
└── project_notes/
├── bugs.md # 50-100 bug entries, solutions
├── decisions.md # 5-15 architectural decisions
├── key_facts.md # Configuration, URLs, facts
└── tracks.md # Work log with ticket referencesWhy docs/project_notes/?
- Standard engineering organization (not AI-specific)
- Clear purpose for all team members
- Coexists with other documentation
File Purposes
| File | Content | Audience | Refresh Rate |
| bugs.md | Bug descriptions, root causes, solutions | Developers | Continuous |
| decisions.md | Architectural decisions with rationale | Architects, leads | Quarterly |
| key_facts.md | Configuration, URLs, ports, facts | All | Monthly |
| tracks.md | Work log, completed tickets | PM, team | Weekly |
Reference Guide
Project Memory Setup & Management
Use when: Creating or maintaining project memory
Covers:
- Initial memory setup and directory structure
- Template content for each memory file
- Configuring AGENT.md with memory protocols
- Searching and updating memory files
- Quarterly refresh process
- Cross-referencing patterns
Quick Start
For New Projects
- Create
docs/project_notes/directory - Create 4 memory files (bugs.md, decisions.md, key_facts.md, tracks.md)
- Add templates from reference guide
- Configure
AGENT.mdwith memory section - Start documenting as you work
For Existing Projects
- Review existing documentation
- Consolidate into 4 memory files
- Add AGENT.md configuration
- Set quarterly refresh reminders
- Migrate as you encounter issues
Memory File Content
bugs.md Structure
### YYYY-MM-DD - Brief Description
- **Issue**: What went wrong
- **Root Cause**: Why it happened
- **Solution**: How it was fixed
- **Prevention**: How to avoid in futureExample:
### 2025-01-25 - Database Connection Timeout
- **Issue**: API requests timing out after 30s
- **Root Cause**: Connection pool default timeout was 30s, too low
- **Solution**: Increased timeout to 300s, added connection validation
- **Prevention**: Monitor connection pool metrics in stagingdecisions.md Structure (ADR Format)
### ADR-XXX: Decision Title (YYYY-MM-DD)
**Context**: Why decision was needed
**Decision**: What was chosen
**Alternatives**: What was rejected and why
**Consequences**: Benefits and trade-offskey_facts.md Structure
## Category Name
- **Item**: Value
- **URL**: Link
- **Port**: Number
- **Config**: Setting
## Database
- Host: prod-db.aws.region
- Port: 5432
- Failover: automatic
## Local Development
- Database: Docker (PostgreSQL)
- Port: 5432
- URL: localhost:5432tracks.md Structure
### YYYY-MM-DD - TICKET-ID: Brief Title
- **Status**: Completed / In Progress / Blocked
- **Description**: What was done
- **URL**: Link to ticket/PR
- **Notes**: Key learnings or blockersIntegration with AGENT.md
Add memory section to AGENT.md:
## Project Memory System
This project maintains institutional knowledge in docs/project_notes/.
### Memory-Aware Protocols
- Before proposing architectural changes: Check decisions.md
- When encountering errors: Search bugs.md for solutions
- When looking up configuration: Check key_facts.md
- When completing work: Log in tracks.mdBest Practices
- ✅ Document as you go: Don't wait until end of week
- ✅ Include rationale: Why, not just what
- ✅ Use dates: Temporal context matters
- ✅ Link references: Cross-reference related entries
- ✅ Never store secrets: Use secure vaults, document access
- ✅ Quarterly refresh: Keep information current
- ✅ Human-readable: Write for team members reading later
Common Patterns
Bug Discovery Workflow
- Encounter bug
- Search bugs.md for similar issues
- If found, apply known solution
- If new, investigate and document
- Log in bugs.md with solution and prevention
Decision Making Workflow
- Face architectural decision
- Check decisions.md for related decisions
- Consider alternatives
- Document in ADR format
- Reference from code/documentation
Configuration Reference
- Need to look up config
- Check key_facts.md
- If not found, document
- Share with team
- Link from AGENT.md
Quarterly Maintenance
Every quarter:
- bugs.md - Archive solved bugs, update documentation
- decisions.md - Review if decisions still valid, update if needed
- key_facts.md - Verify URLs, ports, facts still current
- tracks.md - Archive previous quarter, start fresh
Success Metrics
Good project memory:
- ✅ Team refers to decisions.md before proposing changes
- ✅ Similar bugs don't get re-discovered
- ✅ New team members find onboarding info
- ✅ Configuration is never lost or repeated
- ✅ Work completed is documented for reference
Dependencies
- context-engineer - For context capture and preservation
- state-manager - For state and configuration management
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.

