Solar-Ai
CLAUDE.md1 starsChanged 8 months agoMITView on GitHub
- Reads credentials
In this repository8 other files it gives its agents.
- AGENTS.mdAGENTS.md
- templates/AGENTS.mdAGENTS.md
- vibe-agentsSkill
- vibe-buildSkill
- vibe-prdSkill
- vibe-researchSkill
- vibe-techdesignSkill
- vibe-workflowSkill
The file
Show its rules and text
We split this CLAUDE.md into the rules it gives an agent
13 rules. About 0 are common, 10 are distinctive, and 3 are very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈10
Distinctive · estimate
This author's own choices. Worth a look.
3
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- could be automatic: a setting could enforce it
- Distinctive: line 19
Master Plan First: Always read AGENTS.md before starting any task.
- Distinctive: line 22
Architecture: All LLM calls and DB writes go through the Python agent service.
- Very specific: line 23
Security: Never bypass permission checks.
- Very specific: line 24
Memory: After any milestone or architectural decision, update MEMORY.md.
- Distinctive: line 25
Pre-Commit: Run linters and tests before any commit.
- Distinctive: line 43
Do NOT delete files without explicit confirmation
- Distinctive: line 44
Do NOT modify SQLite schemas without a migration plan
- Distinctive: line 45
Do NOT add features not in the current phase (see AGENTS.md roadmap)
- Distinctive: line 46
Do NOT skip tests for "simple" changes
- Distinctive: line 47
Do NOT bypass failing tests or pre-commit hooks
- Very specific: line 48
Do NOT use any types in TypeScript or untyped Python functions
- Distinctive: line 49
Do NOT store API keys or credentials in source files
- Distinctive: line 50
Do NOT make direct DB calls from the React/TypeScript UI layer
≈ “Common” is an estimate: at least 1 in 5 of the 27,641 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.md — Claude Code Configuration for Solar AI OS
Project Context
App: Solar AI OS Stack: Python (FastAPI) + Tauri + React/TypeScript + SQLite + Claude API + Gmail API Stage: MVP Development — Phase 0 → 1 (Blueprint A, local only) Current Step: Step 5 — Scaffold agent service + SQLite + Claude API + basic Tauri UI
Source Documents (Read Before Each Session)
AGENTS.md— Master plan, active phase, roadmap, agent behavior rulesMEMORY.md— Architectural decisions & session stateagent_docs/tech_stack.md— Full stack details, code patterns, setup commandsagent_docs/product_requirements.md— PRD, user stories, success criteriaagent_docs/testing.md— Test strategy and commandsagent_docs/project_brief.md— Conventions, quality gates, update cadenceREVIEW-CHECKLIST.md— Pre-merge verification checklist
Directives
- Master Plan First: Always read
AGENTS.mdbefore starting any task. It contains the current phase, active tasks, and rules. - Plan Before Coding: Propose a brief step-by-step plan and wait for approval before changing more than one file.
- Incremental Build: Implement one feature at a time. Run tests/linters after each change. Fix failures before moving on.
- Architecture: All LLM calls and DB writes go through the Python agent service. The React UI is display-only — no direct DB access.
- Security: Never bypass permission checks. Never commit
.envor API keys. - Memory: After any milestone or architectural decision, update
MEMORY.md. - Pre-Commit: Run linters and tests before any commit. Fix failures — never skip or bypass.
- Concise: Be brief. Ask one specific clarifying question if critical info is missing.
Commands
# Python agent service
cd agent-service && python main.py # Start service
cd agent-service && pytest # Run tests
cd agent-service && ruff check . # Lint
# Tauri UI
cd ui && npm run tauri dev # Start UI (hot-reload)
cd ui && npm test # Component tests
cd ui && npm run lint # ESLint
cd ui && npm run tauri build # Build distributableWhat NOT To Do
- Do NOT delete files without explicit confirmation
- Do NOT modify SQLite schemas without a migration plan
- Do NOT add features not in the current phase (see AGENTS.md roadmap)
- Do NOT skip tests for "simple" changes
- Do NOT bypass failing tests or pre-commit hooks
- Do NOT use
anytypes in TypeScript or untyped Python functions - Do NOT store API keys or credentials in source files
- Do NOT make direct DB calls from the React/TypeScript UI layer
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.

