vibe-to-prod
CLAUDE.md34 starsChanged 11 months agoMITView on GitHub
Philosophy: Critical rules embedded here. Platform-specific rules in .claude/rules/. Platform-specific rules load automatically when working on relevant code: *Last Updated
In this repository9 other files it gives its agents.
- openspec/AGENTS.mdAGENTS.md
- code-reviewSkill
- continuous-improvementSkill
- dev-workflowSkill
- git-workflowSkill
- openapi-workflowSkill
- ralph-invokeSkill
- security-scanSkill
- systematic-debuggingSkill
The file
We split this CLAUDE.md into the rules it gives an agent
1 rule. About 0 are common, 1 is distinctive, and 0 are very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈1
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
Continuous Improvement: - Every mistake → update rules.
≈ “Common” is an estimate: at least 1 in 5 of the 28,029 CLAUDE.md 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
Claude AI Memory - Vibe to Production
Philosophy: Critical rules embedded here. Platform-specific rules in .claude/rules/.
Core Principles
- Single Source of Truth - Edit source, not derived. OpenAPI → generates code.
- Impact Radius = Verification Radius - Change affects N systems → verify N systems.
- Make Wrong Hard - Automation, types, hooks prevent mistakes.
- Continuous Improvement - Every mistake → update rules. Every task → brief retrospective.
Before ANY Code Change
[ ] What is SOURCE vs DERIVED? (Edit source, regenerate derived)
[ ] What systems are affected? (Verify ALL of them)
[ ] Is there automation? (Check Makefile, scripts/)Critical Rules (All Platforms)
| Rule | Why |
| Use generated types | Clients expect exact JSON from OpenAPI |
| Never edit generated files | PreToolUse hook blocks this automatically |
| Verify build before commit | Backend: make build, iOS: make build, Android: ./gradlew build |
Commit format: type: description | Conventional commits |
Platform Rules (Auto-loaded by path)
Platform-specific rules load automatically when working on relevant code:
| Platform | Rule File | Trigger Path |
| Backend (Go) | .claude/rules/backend.md | backend/** |
| iOS (Swift) | .claude/rules/ios.md | mobile/ios/** |
| Android (Kotlin) | .claude/rules/android.md | mobile/android/** |
| Testing | .claude/rules/testing.md | **/*test* |
Project Structure
.
├── backend/ # Go API server
│ ├── api/openapi.yaml # API spec (SOURCE OF TRUTH)
│ ├── cmd/api/main.go # Entry point
│ └── Makefile
├── mobile/
│ ├── ios/ # Swift/SwiftUI
│ │ ├── project.yml # XcodeGen (SOURCE OF TRUTH)
│ │ └── App/
│ └── android/ # Kotlin/Compose
│ └── app/
├── infrastructure/pulumi/ # IaC (Pulumi + Go)
├── scripts/
│ ├── setup.sh # Initial setup
│ └── openapi_workflow.py # API code generation
└── .claude/ # AI workflow configCommon Workflows
OpenAPI Change
# 1. Edit API spec
backend/api/openapi.yaml
# 2. Regenerate everything
python scripts/openapi_workflow.py --full
# 3. Verify ALL platforms build
make verify-all # or run each: backend, iOS, AndroidNew Backend Endpoint
# 1. Add to OpenAPI spec
# 2. Regenerate: cd backend && make generate
# 3. Implement handler
# 4. Add tests
# 5. Verify: make build && make testAdding a Feature
# 1. Plan with /ultra-think if complex
# 2. Implement with platform rules loaded
# 3. Test across affected platforms
# 4. Review with /code-review
# 5. Commit with /commitSlash Commands
| Command | Purpose |
/commit | Create well-formatted commit |
/code-review | Comprehensive code review |
/generate-tests | Generate test suite |
/refactor-code | Safe refactoring |
/ultra-think | Deep analysis mode |
/security-scan | Security vulnerability check |
/improve-claude-config | Audit and improve config |
Configuration
| File | Purpose |
.claude/settings.json | Team-shared settings (hooks, permissions) |
.claude/settings.local.json | Personal overrides (gitignored) |
.claude/rules/ | Path-conditional rules |
.claude/skills/ | Model-invoked skills |
.claude/commands/ | User-invoked slash commands |
.claude/hooks/ | Automation hooks |
.claude/CONFIGURATION_IMPROVEMENT_GUIDE.md | Config best practices |
Reference Docs
| Topic | Location |
| Getting started | docs/GETTING_STARTED.md |
| Architecture decisions | docs/ARCHITECTURE.md |
| API gateway/auth | docs/API_GATEWAY.md |
| CI/CD setup | docs/CICD.md |
| MCP servers | docs/AI_WORKFLOW.md |
| Troubleshooting | docs/TROUBLESHOOTING.md |
Last Updated: 2026-01-02 Version: 1.0.0 - Initial AI memory setup
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.

