pentestagent
GH05TCREW/pentestagent/CLAUDE.md
support any provider (Anthropic, OpenAI, etc.) and exposes a TUI, a CLI, and an MCP server interface. The agent can run tools locally or inside a Docker sandbox (base
How real projects brief Claude Code: the commands, conventions and warnings they put in CLAUDE.md.
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.
GH05TCREW/pentestagent/CLAUDE.md
support any provider (Anthropic, OpenAI, etc.) and exposes a TUI, a CLI, and an MCP server interface. The agent can run tools locally or inside a Docker sandbox (base
evinjohnn/natively-cluely-ai-assistant/CLAUDE.md
graph **IMPORTANT: This project has a knowledge graph. ALWAYS use the `code-review-graph` MCP tools BEFORE using Grep, Glob, Read, or broad filesystem exploration.** The graph is faster, cheaper
blazickjp/arxiv-mcp-server/CLAUDE.md
CLAUDE.md Guidance for coding agents working in this repository. ## Project `arxiv-mcp-server` is a Python 3.11+ Model Context Protocol server. It exposes arXiv search, metadata, paper download/read, original LaTeX
Chachamaru127/claude-code-harness/CLAUDE.md
templates/` Templates / `tests/` Validation / `go/` Harness v4 Go native engine ([SPEC.md](go/SPEC.md), [DESIGN.md](go/DESIGN.md)) / `mcp-server/` MCP server implementation / `harness-ui/` UI subproject / `opencode/` OpenCode-compatible output ## Using Skills (Important
punitarani/fli/CLAUDE.md
interface** (`fli/cli/`) - Typer-based command line tool with `flights` and `dates` commands - **MCP server** (`fli/mcp/`) - Model Context Protocol server for AI assistant integration - **Core utilities** (`fli/core/`) - Shared parsing and building
coder/claudecode.nvim/CLAUDE.md
lua/claudecode/types.lua` `ClaudeCodeSendCompleteData` and README "Events". ## MCP Protocol Compliance ### Protocol Implementation Status - ✅ **WebSocket Server**: RFC 6455 compliant with MCP message format - ✅ **Tool Registration**: JSON Schema-based tool definitions - ✅ **Authentication
Steady5 mentions in 2 weeks
No other project uses it yet
SynkraAI/aios-core/.claude/CLAUDE.md
needed - Limit tool search to maximum 2 searches per turn to avoid overhead - Essential MCP servers (nogic, code-graph) are never disabled - Non-essential servers (EXA, Apify, Playwright) load
supermemoryai/apple-mcp/CLAUDE.md
apple-mcp Development Guidelines ## Commands - `bun run dev` - Start the development server - No specific test or lint commands defined in package.json ## Code Style ### TypeScript Configuration - Target: ESNext - Module: ESNext - Strict
remix-project-org/remix-project/CLAUDE.md
remix-solidity/ # Compiler management ├── remix-tests/ # Solidity unit testing ├── remix-ai-core/ # AI features & MCP server ⭐ ├── remix-core-plugin/ # Plugin base classes ├── remix-ui/ # React component library (many sub-packages
sveltia/sveltia-cms/CLAUDE.md
import`. - Svelte 5: runes syntax only — no legacy Svelte patterns. Use the Svelte MCP server / `svelte-file-editor` agent for any `.svelte` or `.svelte.js`/`.svelte.ts` work. - Prose: Canadian English
Steady21 mentions in 2 weeks
No other project uses it yet
allenpeng0705/EnvoyMesh/CLAUDE.md
signs mandates & device certificates | | **Device** | `envoy:device:<sha256(pubkey)>` | A specific device (laptop, phone, server), authorized by owner | | **Agent** | `envoy:agent:<sha256(ownerId + agent-pubkey)>` | AI agent running on owner
vectorize-io/self-driving-agents/website/CLAUDE.md
page sells **how that runtime gains self-driving agents** — not how a plugin or MCP server works. ## Per-harness page structure Every harness page (`/harnesses/ `) follows the same three-section
lowdefy/lowdefy/CLAUDE.md
writes .mcp.json/AGENTS.md/skill into a project; its `.mcp.json` entry runs `lowdefy mcp`, a stdio MCP server that routes each tool call to the right app/worktree's dev server and starts
apptension/saas-boilerplate/CLAUDE.md
packages/webapp`, `packages/webapp-libs`); Django on Python `>=3.11`, uv-managed (`packages/backend`); Celery (`packages/workers`); AWS CDK (`packages/infra`); MCP server (`packages/mcp-server`) | | Verification | Per project via Nx: `pnpm nx run :lint`, `:test`, `:build`, `:type-check
marketcalls/openalgo/CLAUDE.md
cannot set custom HTTP headers. This is an accepted architectural trade-off. - The stdio MCP server (`mcp/mcpserver.py`) is local-only and not remotely exposed. `blueprints/mcp_http.py` and `blueprints/mcp_oauth.py` are the remote
cocoindex-io/cocoindex-code/CLAUDE.md
that accepts connections over a Unix socket (Windows: named pipe). The `ccc` CLI and MCP server communicate with it via short-lived **per-request connections** (`client.py`): connect → handshake → send
Steady3 mentions in 2 weeks
No other project uses it yet
event-catalog/eventcatalog/CLAUDE.md
canvas (`@eventcatalog/visualiser/studio`) And one app: - `@eventcatalog/studio-web` (`apps/studio-web`, MIT) - Studio on the web: a public MCP server and canvas (no sign-in, stores nothing), deployed on its own ## Quick Reference Commands
Steady2 mentions in 2 weeks
No other project uses it yet
VibiumDev/vibium/CLAUDE.md
Tech Stack - Go (vibium binary) - TypeScript (JS client) - Python (Python client) - WebDriver BiDi protocol - MCP server (stdio) ## Design Philosophy Optimize for first-time user/developer joy. Defaults should create
No other project uses it yet
F1bonacc1/process-compose/CLAUDE.md
more. It operates in multiple modes: TUI (default), HTTP API server, daemon, and MCP server. ## Build & Development Commands ```bash make build # Build binary to bin/process-compose (CGO_ENABLED=0) make test
No other project uses it yet
anthropics/commerce-agents/CLAUDE.md
console. - `*/managed-agents/`: the manifest directory (with the derived `system.md`) and the role's MCP server. - `examples/demo_common/` and `examples/web-shared/`: what the verticals' APIs and web apps share; `examples
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.