How real projects brief Claude Code: the commands, conventions and warnings they put in CLAUDE.md.
wednesday-solutionsCLAUDE.mdwednesday-solutions/ai-agent-skills/CLAUDE.md
projects. Research agent runs first to build domain context, then Architect, PM, and Security agents run in parallel. Synthesis agent combines all perspectives into a detailed GSD-style PLAN.md with
wellwelwelCLAUDE.mdwellwelwel/lagune/CLAUDE.md
Security-Driven Hardening): Lagune > This file orients any AI agent (and human) working in this repository. Read it fully before making changes. It describes **what Lagune
huiferCLAUDE.mdhuifer/skill-security-scan/CLAUDE.md
Claude Code (claude.ai/code) when working with code in this repository. ## Project Overview **skill-security-scan** is a command-line security scanner for Claude Skills. It performs static analysis
MatanYeminiCLAUDE.mdMatanYemini/bitbucket-mcp/CLAUDE.md
used to prevent accidental data loss. Every PR is analyzed with CodeQL for security. ## Development Commands ### Build and Run ```bash npm install # Install dependencies npm run build # Compile TypeScript
paimlCLAUDE.mdpaiml/paiml-mcp-agent-toolkit/CLAUDE.md
gate`.** Its `--checks` argument accepts only: dead-code, complexity, coverage, sections, provability, satd, entropy, security, duplicates, all. Passing docs-accuracy exits 2 with a clap error. `pmat validate-readme` above
alexei-ledCLAUDE.mdalexei-led/spotinfo/CLAUDE.md
part of a routine dependency bump; revisit only if v2 stops getting security fixes. - `github.com/jedib0t/go-pretty/v6` - Table formatting - `github.com/aws/aws-sdk-go-v2` - AWS SDK for live pricing and placement scores - `golang.org/x/net/html
antonbabenkoCLAUDE.mdantonbabenko/deliberation/CLAUDE.md
specialized expert subagents. Seven domain experts: Architect, Plan Reviewer, Scope Analyst, Code Reviewer, Security Analyst, Researcher, and Debugger. Only Gemini can advise OR implement. (GPT, Grok, and OpenRouter are advisory
Arman-KudaibergenovCLAUDE.mdArman-Kudaibergenov/1c-ai-development-kit/CLAUDE.md
Read - Grep - Glob - Bash --- # Code Review Review agent work before merge. Goal: catch bugs, security issues, and scope creep. Keep feedback actionable with file:line references. ## Step
mtarcureCLAUDE.mdmtarcure/claude-vibe-squad/chrono/CLAUDE.md
same word for both turns a cosmetic issue into what sounds like a security incident, and spends the operator's attention on alarm rather than judgement. ### Evidence freshness Any measurement
liangdabiaoCLAUDE.mdliangdabiao/autogen-financial-analysis/CLAUDE.md
Supporting Infrastructure**: - `PerformanceOptimizer` - Async performance optimization - `SystemMonitor` - Prometheus-based monitoring - `SecurityManager` - Data encryption and security - `ErrorHandler` - Advanced error handling with retry logic ## Development Commands This is a documentation/learning project with
jaccenCLAUDE.mdjaccen/Awesome-Gaussian-Skills/CLAUDE.md
3D Spatial Intelligence Open-Source Toolbox for 3D Gaussian Splatting Research. 872 methods knowledge base, 15 research-grade skills (3 Router architecture), interactive explorer. Covers 3DGS paper reading, method comparison, code review, experiment planning, CAD/Mesh bridge, visualization, NeRF migration, engineering deployment, CG paper writing, IP generation, spatial intelligence, MCP rendering (spec-first sculpting + code-first export), articulated reasoning, compression & deployment, training debugging. SLAT unified representation framework for conversion skills.
Open-Document-AllianceCLAUDE.mdOpen-Document-Alliance/PDF-Tools/CLAUDE.md
Guide users to correct tools (e.g., "use 'read_pdf_fields' to see available fields") ### Security - Never log or expose passwords - Validate file paths to prevent directory traversal - Use secure defaults
smerchekCLAUDE.mdsmerchek/claude-epub-skill/CLAUDE.md
existing functionality - `deprecated` - Soon-to-be removed features - `removed` - Removed features - `fixed` - Bug fixes - `security` - Security improvements #### 2. Prepare Release When ready to release, prepare the changelog: ```bash # Move unreleased
raullenchaiCLAUDE.mdraullenchai/vnsh/CLAUDE.md
Analytics Engine binding `VNSH_ANALYTICS`. Secrets belong in Wrangler secrets, never source or `wrangler.toml`. ## Security invariants - Never send fragment keys to the server or place them in logs. - Never infer
TradersPostCLAUDE.mdTradersPost/pinescript-agents/CLAUDE.md
Development utilities ## Essential Limits - Max 500 bars lookback - Max 500 plot outputs - Max 40 security() calls - Max 64 drawing objects - Strings: 40,960 characters ## Best Practices 1. **Version Declaration**: Always
GenAI-Security-ProjectCLAUDE.mdGenAI-Security-Project/agent-control-standard/CLAUDE.md
STYLE.md. ## Project Overview ACS (Agent Control Standard) is the industry standard for building secure, observable AI agents. It delivers three core capabilities: - **Inspectability**: Complete visibility into agent components and capabilities
paulpreibischCLAUDE.mdpaulpreibisch/AgentVibes/CLAUDE.md
dirty tree. - Alpha flow is normal: ship `x.y.z-alpha.N` while iterating, then promote to stable. ## Security (project-specific, enforced by tests) - **User-supplied file paths** (custom music, config) go through
buckle42CLAUDE.mdbuckle42/claude-code-remote/CLAUDE.md
basic auth is not currently enabled. It was causing connection failures. Tailscale network-level security is the primary access control. - **Apple Silicon vs Intel:** tmux-attach.sh uses `which tmux` to auto
MicrosoftCLAUDE.mdmicrosoft/simplechat/CLAUDE.md
class instead - Use Bootstrap alert classes for notifications, not `alert()` calls ## Route Decorators — Swagger Security **Every Flask route MUST include the `@swagger_route(security=get_auth_security())` decorator.** - Import `swagger
qufei1993CLAUDE.mdqufei1993/cc-weixin/CLAUDE.md
/marketplace.json` - Test files use `WEIXIN_STATE_DIR` env var to isolate from real credentials ## Security Rules - Never log or expose tokens, aes_keys, or account credentials - Never commit `.env`, `account.json
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.
