CLAUDE.md examples from real projects

How real projects brief Claude Code: the commands, conventions and warnings they put in CLAUDE.md.

Best matches · from page 7Worked for most · soon

Asymptote-LabsCLAUDE.md

agent-beacon

Asymptote-Labs/agent-beacon/CLAUDE.md

forwarding patterns and the opt-in Asymptote Managed path. - Dependency vulnerability scanning or package security remediation. ## Common Commands Run tests for the public CLI. Build the embedded hooks binary first

1.6k6d agoReads credentialsDiscuss

actionbookCLAUDE.md

actionbook

actionbook/actionbook/CLAUDE.md

fields — otherwise concurrent sessions produce indistinguishable logs. Level controlled via `RUST_LOG` or `--verbose`. --- ### Security & Feature Boundaries - **CDP method security levels.** L1: read-only (screenshot, DOM read). L2: modification (click

1.6k2mo agoReads credentialsDiscuss

DockerCLAUDE.md

mcp-gateway

docker/mcp-gateway/CLAUDE.md

Overview This is the **Docker MCP Gateway** - a CLI plugin that enables easy and secure running of Model Context Protocol (MCP) servers through Docker containers. The plugin acts

1.6k9mo agoDiscuss

rivet-devCLAUDE.md

sandbox-agent / foundry

rivet-dev/sandbox-agent/foundry/CLAUDE.md

swaps when a different user sends a message. See `.context/proposal-task-owner-git-auth.md` for the full design. - **Security: OAuth token scope.** The user's GitHub OAuth token has `repo` scope, granting full control

1.6k7mo agoReads credentialsDiscuss

fynnflueggeCLAUDE.md

agtx

fynnfluegge/agtx/CLAUDE.md

paths exactly (agtx seeds each worktree into `trustedWorkspaces` when the root is trusted). **`AgentDialog::security` splits the table** — trust/bypass prompts are the user's decision: with `auto_trust = false` (default

1.5k5mo agoDiscuss

silversteinCLAUDE.md

minutes

silverstein/minutes/CLAUDE.md

capture + streaming) │ │ ├── transcribe.rs # Batch transcription: whisper.cpp. Retained Parakeet config resolves to Whisper until secure transport exists │ │ ├── diarize.rs # Speaker diarization + attribution types (pyannote-rs, or energy-based from per-source stems

1.5k28d agoDiscuss

Open-Source-LegalCLAUDE.md

OpenContracts

Open-Source-Legal/OpenContracts/CLAUDE.md

issue #820): operational exceptions are caught and returned as error strings to the LLM; security exceptions (`PermissionError`, `ToolConfirmationRequired`) propagate. - Pre-execution checks run on every call (not cached): permission validation

1.5k12mo agoDiscuss

umputunCLAUDE.md

ralphex

umputun/ralphex/CLAUDE.md

# ralphex Autonomous plan execution with Claude Code - Go rewrite of ralph.py. ## LLM Documentation See

1.5k41d agoReads credentialsDiscuss

ghostwrightCLAUDE.md

phantom

ghostwright/phantom/CLAUDE.md

docker-entrypoint.sh # Docker bootstrap (wait for deps, model pull, init) docs/ # Documentation (architecture, channels, mcp, security, etc.) ``` ## Architecture Overview Message flow: Slack message -> SlackChannel adapter -> ChannelRouter -> SessionManager (find/create session) -> PromptAssembler (base

1.5k6mo agoReads credentialsDiscuss

win4rCLAUDE.md

ClawTeam-OpenClaw

win4r/ClawTeam-OpenClaw/CLAUDE.md

DATA_DIR` relocates team/task/inbox state but NOT global config. ## Cross-cutting conventions ### Path discipline (security-critical) Every user-supplied name (team, agent, user, plan ID, inbox recipient…) must flow through

1.4k6mo agoDiscuss

daymadeCLAUDE.md

claude-code-skills

daymade/claude-code-skills/CLAUDE.md

work). Treat [packaging_policy.py](daymade-skill/skill-creator/scripts/packaging_policy.py) as the canonical inclusion policy for packaging, security attestation, source audits, and version checks. Keep consumers on this shared implementation. Preserve the recorded policy

1.4k6mo agoPipes a download into a shellDiscuss

CloudAI-XCLAUDE.md

claude-workflow-v2

CloudAI-X/claude-workflow-v2/CLAUDE.md

1.4k9mo agoDiscuss

tigerless-labsCLAUDE.md

agent-memory

tigerless-labs/agent-memory/CLAUDE.md

behavior primes it. Reserve explicit prohibitions for failure modes actually observed in traces. ## Security — agents run autonomously, so the system must fail safe - **Trust nothing by default.** Treat external input

1.4k29d agoDeletes or force-pushesDiscuss

nrslibCLAUDE.md

takt

nrslib/takt/CLAUDE.md

fallback-heavy logic. TAKT is a local tool: no audit trails, tamper-resistance, or security theater. - Filenames mostly `kebab-case`. Conventional Commit style with occasional `(#issue)` suffix. - Don't commit

1.4k16d agoReads credentialsDiscuss

m0n0x41dCLAUDE.md

haft

m0n0x41d/haft/CLAUDE.md

execution authority; approve, reopen, or rebaseline a SpecSection; make a material product/value/scope, public-promise, security, legal, privacy, finance, irreversible-data, compatibility, or authority-allocation choice. - If current Work would rely

1.4k50d agoDiscuss

peterkrueckCLAUDE.md

Claude-Code-Development-Kit / templates

peterkrueck/Claude-Code-Development-Kit/templates/CLAUDE.md

keep docs in sync with the code. ## 6. Coding Standards ## 7. Testing ## 8. Privacy & Security

1.4k4mo agoDiscuss

elirantutiaCLAUDE.md

vibeyard

elirantutia/vibeyard/CLAUDE.md

store.ts`. CLI tool behavior is abstracted via the provider system (`src/main/providers/`). - **Preload** (`src/preload/preload.ts`) — Secure bridge exposing `window.vibeyard` API via `contextBridge` with namespaces: `pty`, `session`, `store`, `profiles`, `fs`, `provider`, `menu`. - **Renderer

1.4k4mo agoReads credentialsDiscuss

utkusenCLAUDE.md

sast-skills / sast-files

utkusen/sast-skills/sast-files/CLAUDE.md

SAST Security Assessment Your goal is to identify security vulnerabilities in the codebase located in the current directory. --- ## Step 1: Codebase Analysis & Threat Modeling Before running, check if `sast/architecture.md` already

1.3k6mo agoDiscuss

greyhaven-aiCLAUDE.md

autocontext

greyhaven-ai/autocontext/CLAUDE.md

ENABLED` (opt-in persistent interpreter workspace for multi-generation runs; lifecycle isolation, not a security sandbox), `AUTOCONTEXT_WORKSPACE_INTERPRETER_TIMEOUT_SECONDS` (see `config/workspace_interpreter.py`) - **Judge**: `AUTOCONTEXT_JUDGE_PROVIDER`, `AUTOCONTEXT_JUDGE_MODEL

1.3k7mo agoReads credentialsDiscuss

sipyourdrink-ltdCLAUDE.md

bernstein

sipyourdrink-ltd/bernstein/CLAUDE.md

1.3k9d agoDiscuss

CLAUDE.md vs AGENTS.md

About CLAUDE.md

What is CLAUDE.md?

A file Claude Code reads at the start of every session. It holds the commands, conventions and warnings the agent needs for this project.

Where does it go?

At the repository root. Claude Code also reads CLAUDE.md files in subdirectories when it works there.

What should it contain?

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.md or AGENTS.md?

Claude Code reads CLAUDE.md; most other agents read AGENTS.md. Many projects keep one and point the other at it.