How real projects brief Claude Code: the commands, conventions and warnings they put in CLAUDE.md.
bybren-llcCLAUDE.mdbybren-llc/safe-agentic-workflow/CLAUDE.md
workflow) ├── specs/ # SAFe specifications (Epic/Feature/Story) ├── patterns_library/ # Reusable code patterns (7 categories) ├── .claude/ # Claude Code harness (hooks, commands, skills, agents) ├── agent_providers/ # Agent configurations └── scripts/ # Utility scripts ``` --- ## SAFe Workflow
LangChainCLAUDE.mdlangchain-ai/langchain-google/CLAUDE.md
test file uv run --group test pytest tests/unit_tests/test_specific.py ``` ```bash # Lint code make lint # Format code make format # Type checking uv run --group lint mypy . ``` #### Key config files - pyproject.toml: Main workspace
tobihagemannCLAUDE.mdtobihagemann/turbo/claude/CLAUDE.md
changed files and report findings directly, the way `/create-skill` reviews a skill, without invoking `$review-code` or any other skill. Review skills carry user-input gates that `codex exec` drops
jamditisCLAUDE.mdjamditis/claude-skills-journalism/CLAUDE.md
written implementation plan ├── finishing-a-development-branch/ # Decide how to land completed work ├── receiving-code-review/ # Handle review feedback ├── requesting-code-review/ # Verify completed work before merging ├── subagent-driven-development
carloslfuCLAUDE.mdcarloslfu/slotstream/CLAUDE.md
# slotstream — agent instructions Qwen3.8-Flash-Next (125B + 51B n-gram, 4-bit, ~104 GB) on Macs
eriklangilleCLAUDE.mderiklangille/clauntty/CLAUDE.md
Wait for build processing (5-30 min) 4. Add testers (internal = instant, external = requires review) ## Logging & Debugging ### Log Levels The app uses a tiered logging system optimized for performance: | Method
nudgebeeCLAUDE.mdnudgebee/nudgebee/app/CLAUDE.md
surface in tests or manual QA of the tab itself. Any AI review pass (Gemini Code Assist, `/create-pr` self-review, `/review-pr`) MUST flag a diff that adds/changes a tab `fragment
nudgebeeCLAUDE.mdnudgebee/nudgebee/CLAUDE.md
service entry points. This file (`CLAUDE.md`) is primarily guidance for AI coding agents (Claude Code, etc.), but a few sections are worth a human contributor's time: > > - **[Architecture Decisions](docs/architecture-decisions.md
sonichiCLAUDE.mdsonichi/sutando/CLAUDE.md
machine-readable `checks:` block (hardcoded-path scan) in CI; and Claude Code's managed GitHub-App reviewer reads `REVIEW.md` directly. Adding or editing a lesson is a PR to `REVIEW.md
zhuyansenCLAUDE.mdzhuyansen/agent-skills-hub/CLAUDE.md
migration without documenting the security impact. ### Forbidden Phrases in Code These phrases in comments or commit messages trigger mandatory review: - "should work now" → must include test evidence - "trivial change" → must
owainlewisCLAUDE.mdowainlewis/blueprint/CLAUDE.md
/improve` for one kind of engineering work. Use `/task-to-pr` to deliver one or more code changes. Use `/codex-issue-coordinator` only in Codex to coordinate a large GitHub issue batch through separate
brainqub3CLAUDE.mdbrainqub3/RLM/CLAUDE.md
codex-headless.ps1 "review the current repo and list the highest-risk files" .\codex-headless.cmd "review the current repo and list the highest-risk files" ``` The wrapper runs `codex exec` from
syahiidkamilCLAUDE.mdsyahiidkamil/Software-Engineer-AI-Agent-Atlas/CLAUDE.md
Visualize architecture when clarity helps 5. **Don't reinvent the wheel** - Before writing new code, I check whether it already exists in the codebase or is solved by a well
smithersaiCLAUDE.mdsmithersai/claude-p/CLAUDE.md
spec in the same change. 2. **TDD.** Write the failing test, then the code. The test order documented in §6 of `SPEC.md` is the implementation order. Don't skip ahead
GRCEngClubCLAUDE.mdGRCEngClub/claude-grc-engineering/CLAUDE.md
NIST ``` **2. Implementation Phase**: ```bash # Generate production-ready code /grc-engineer:generate-implementation access_control_account_management aws ./output # Review and customize cd ./output/terraform terraform init terraform plan ``` **3. Validation Phase
m98CLAUDE.mdm98/fluent/CLAUDE.md
Primary Role: Interactive Language Tutor You are a personal language tutor, powered by Claude Code. Your mission is to help learners master their target language through **fun, interactive, systematic learning
MongoDBCLAUDE.mdmongodb/mongo-kafka/CLAUDE.md
CLAUDE.md This file provides guidance to Claude Code (claude.ai/code) when working with code in this repository. ## Project Official MongoDB Kafka Connector — a Kafka Connect plugin that ships both
therealarthurCLAUDE.mdtherealarthur/myrlin-workbook/CLAUDE.md
feature: `git worktree add -b feat/ ../cwm-feat- ` 2. **Spawn a new Claude Code session** pointed at the worktree directory 3. **Feed it ALL relevant context** — point it at every file
ombulabsCLAUDE.mdombulabs/claude-code_rails-upgrade-skill/CLAUDE.md
code via the per-pattern `fix:` field, not via `dependencies:`. `bin/validate-patterns` does not enforce the rules in this section — they are editorial guidance for authors and reviewers. The validator only
mrwoguCLAUDE.mdmrwogu/promptscript/CLAUDE.md
Parser: Chevrotain - CLI: Commander.js - Testing: Vitest - Linting: ESLint + Prettier - Project: PromptScript ## Engineering Standards - Use code-review-graph MCP tools BEFORE Grep/Glob/Read when exploring the codebase - Exploring code: use semantic_search
A file Claude Code reads at the start of every session. It holds the commands, conventions and warnings the agent needs for this project.
At the repository root. Claude Code also reads CLAUDE.md files in subdirectories when it works there.
Build and test commands, the project's layout, conventions that aren't obvious from the code, and mistakes to avoid. Short files tend to work better than long ones.
Claude Code reads CLAUDE.md; most other agents read AGENTS.md. Many projects keep one and point the other at it.
