How real projects brief Claude Code: the commands, conventions and warnings they put in CLAUDE.md.
bassimeledathCLAUDE.mdbassimeledath/dispatch/CLAUDE.md
setup, symlink structure, CI | [docs/development.md](docs/development.md) | ## Conventions - **Task IDs**: Short, descriptive, kebab-case (`security-review`, `add-auth`, `fix-login-bug`). - **Checklist markers**: `[x]` done, `[ ]` pending, `[?]` blocked, `[!]` error. - **`.dispatch/` directory**: Ephemeral
dsifryCLAUDE.mddsifry/metaswarm/CLAUDE.md
design-review-gate` skill 3. **WAIT** for all 5 review agents (PM, Architect, Designer, Security, CTO) to approve 4. **ONLY THEN** proceed to planning/implementation This is mandatory even
ruilisiCLAUDE.mdruilisi/lsbot/CLAUDE.md
current branch compared to main. Provide structured feedback on correctness, design, performance, security, and test coverage. ### /github - GitHub CLI Reference Use the `gh` CLI to interact with GitHub. Always specify
tkaufmannCLAUDE.mdtkaufmann/claude-gemini-bridge/CLAUDE.md
calls for replay - Use `DRY_RUN=true` to test delegation logic without calling Gemini ## Security Considerations ### File Exclusions The bridge automatically excludes sensitive files: - `*.secret`, `*.key`, `*.env` - `*.password`, `*.token
bybren-llcCLAUDE.mdbybren-llc/safe-agentic-workflow/CLAUDE.md
Agent team quick reference ├── CONTRIBUTING.md # Git workflow and commit standards ├── docs/ # Documentation (onboarding, database, security, sop, workflow) ├── specs/ # SAFe specifications (Epic/Feature/Story) ├── patterns_library/ # Reusable code patterns (7 categories) ├── .claude/ # Claude
LangChainCLAUDE.mdlangchain-ai/langchain-google/CLAUDE.md
flaky tests) - [ ] Does the test suite fail if your new logic is broken? ### Security and risk assessment - No `eval()`, `exec()`, or `pickle` on user-controlled input - Proper exception handling
jamditisCLAUDE.mdjamditis/claude-skills-journalism/CLAUDE.md
instructions, evidence, and state │ ├── director/ # Explicit top-tier direction and delegation │ ├── electron-dev/ # Electron security model, packaging │ ├── mobile-debugging/ # Eruda, vConsole, remote debug │ ├── one-way-door/ # Block irreversible decisions │ ├── python
zxdxjtuCLAUDE.mdzxdxjtu/claudecode-rule2hook/CLAUDE.md
modifying files in the tests/ directory - Clear build cache when .env file is modified ## Security - Scan for hardcoded API keys before saving configuration files - Validate environment variables before running deployment
aws-solutions-library-samplesCLAUDE.mdaws-solutions-library-samples/guidance-for-claude-code-with-amazon-bedrock/CLAUDE.md
guards - `deployment/CLAUDE.md` — CloudFormation templates, IAM, conditions - `deployment/infrastructure/lambda-functions/CLAUDE.md` — Lambda auth, response patterns - `docs/CLAUDE.md` — Documentation naming, security, structure ## Key Principles 1. **Auth path coverage:** Features touching identity must work across OIDC
aws-solutions-library-samplesCLAUDE.mdaws-solutions-library-samples/guidance-for-claude-code-with-amazon-bedrock/docs/CLAUDE.md
inline. - If content exists in another doc, link to it — don't copy. ### Security in Docs - **Never** include real AWS account IDs, internal URLs, or API keys in examples
sonichiCLAUDE.mdsonichi/sutando/CLAUDE.md
# Sutando You are operating as part of Sutando — a personal AI agent that belongs
zhuyansenCLAUDE.mdzhuyansen/agent-skills-hub/CLAUDE.md
test with single email first. - **Never modify RLS policies** via migration without documenting the security impact. ### Forbidden Phrases in Code These phrases in comments or commit messages trigger mandatory review
GRCEngClubCLAUDE.mdGRCEngClub/claude-grc-engineering/CLAUDE.md
control with framework mappings: ```yaml controls: my_new_control: name: Descriptive Control Name category: Security Category nist_800_53: [AC-X, AC-X(1)] iso_27001: [A.X.Y.Z] soc2: [CC.X
MongoDBCLAUDE.mdmongodb/mongo-kafka/CLAUDE.md
alongside Maven publish. The `dependencies` block in `build.gradle.kts` carries several `TODO: Remove this override` security pins (CVE patches for transitive deps via Checkstyle, SpotBugs, Avro, and Kafka). When bumping those
ombulabsCLAUDE.mdombulabs/claude-code_rails-upgrade-skill/CLAUDE.md
ApplicationRecord`, removed DSL options that raise). - **Silently wrong behavior with production impact** — data loss, security regression, broken auth/CSRF, or cache key mismatches that invalidate stored data. If the user cannot
philippbCLAUDE.mdphilippb/claude-homeassistant/CLAUDE.md
input` tags which are normal and expected - **Secrets are skipped** during validation for security - **SSH access required** for pull/push operations - **Python venv required** for validation tools - All python tools need
mrwoguCLAUDE.mdmrwogu/promptscript/CLAUDE.md
/quality - Review code for quality improvements /refactor - Suggest refactoring opportunities /security-review - Review code for security vulnerabilities /threat-model - Analyze potential security threats /cli - Create CLI command handler /export - Design public
hypnguyen1209CLAUDE.mdhypnguyen1209/offensive-claude/CLAUDE.md
Security Research & Offensive Operations Config You are a specialized AI assistant for advanced security research, vulnerability analysis, and offensive red team operations. You operate with deep expertise in: - **Offensive Security
smart-mcp-proxyCLAUDE.mdsmart-mcp-proxy/mcpproxy-go/CLAUDE.md
Model Context Protocol (MCP): intelligent tool discovery, massive token savings, and built-in security quarantine against malicious MCP servers. **Stack**: Go 1.26 (backend) · TypeScript 5.9 / Vue 3.5 (frontend) · Swift
superbasicstudioCLAUDE.mdsuperbasicstudio/claude-conductor/CLAUDE.md
framework includes a systematic error ledger (`CONDUCTOR.md:241-275`): - **P0**: Complete outage, data loss, security breach - **P1**: Major functionality broken - Error IDs: `ERR-YYYY-MM-DD-001` - Linked to JOURNAL.md entries
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.
