pikiloom
xiaotonng/pikiloom/CLAUDE.md
proxy. Providers + Profiles vault injects credentials per agent at spawn time. 4. **Tool** — Skills, MCP servers, CLI tools, merged across global / workspace scopes. The orchestrator is the product. Lead with
How real projects brief Claude Code: the commands, conventions and warnings they put in CLAUDE.md.
xiaotonng/pikiloom/CLAUDE.md
proxy. Providers + Profiles vault injects credentials per agent at spawn time. 4. **Tool** — Skills, MCP servers, CLI tools, merged across global / workspace scopes. The orchestrator is the product. Lead with
HarmonicSecurity/claudit-sec/CLAUDE.md
Settings | `claude_desktop_config.json` → preferences | keepAwakeEnabled, allowAllBrowserActions, menuBar, sidebar | | Cowork Settings | `cowork_settings.json`, `config.json` | scheduledTasks, webSearch, networkMode | | MCP Servers | `claude_desktop_config.json` → mcpServers | Server names, commands, env var keys | | Plugins | `installed_plugins.json`, remote manifest, marketplace cache
varun29ankuS/shodh-memory/CLAUDE.md
this codebase ## Codebase Map - `src/` — Rust core (memory, API server, embeddings, graph, vector search) - `mcp-server/` — TypeScript MCP server (45 tools, index.ts) - `tui/` — Rust TUI dashboard - `hooks/` — Claude Code hooks
Higangssh/homebutler/CLAUDE.md
binary for running a small home server: CLI, TUI, web dashboard, and an MCP server, with no database and nothing running on the machines it watches. ## Verify before claiming anything
FootprintAI/Containarium/CLAUDE.md
OpenAPI viewer) — they cannot drift because they all regenerate from the same source. The MCP server (which speaks REST through grpc-gateway) gets every new endpoint for free
sudocode-ai/sudocode/.claude/CLAUDE.md
types/ # Shared TypeScript definitions (build first) ├── cli/ # CLI (@sudocode-ai/cli) - core operations ├── mcp/ # MCP server (@sudocode-ai/mcp) - wraps CLI ├── server/ # Local backend (@sudocode-ai/local-server) - executions + API ├── frontend
DitriXNew/EDT-MCP/CLAUDE.md
code conduct (the minefield map) A 1C:EDT plugin (Maven/Tycho, Eclipse OSGi) exposing an MCP server that drives EDT. This file is **what NOT to do** and **where to stop
opensesh/KARIMO/CLAUDE.md
replace GitHub Projects for tracking - Branch naming: `worktree/{prd-slug}-{task-id}` **Requirements:** - GitHub MCP server configured in Claude Code - gh CLI authenticated with `repo` scope **Benefits:** - Complete traceability (task
testdouble/han/CLAUDE.md
coding` because its wrapper skills run skills from each, requires a configured Atlassian MCP server, and is likewise _not_ bundled by the `han` meta-plugin), `han-linear
loulanyue/awesome-claude-notes/CLAUDE.md
persistence, pre/post-tool hooks) - **rules/** - Always-follow guidelines (security, coding style, testing requirements) - **mcp-configs/** - MCP server configurations for external integrations - **scripts/** - Cross-platform Node.js utilities for hooks and setup - **tests
AkaraChen/aghub/CLAUDE.md
Claude, OpenCode, Cursor, Windsurf, Copilot, RooCode, Cline, Gemini, Codex, Zed, Warp, and more), handling MCP servers and skills through a unified interface. Full agent list: `crates/core/src/models.rs` or `aghub-cli --help
speakeasy-api/gram/CLAUDE.md
Frontend React application for Gram. Gram Elements — a chat interface that integrates with Gram MCP servers — lives inside it at `client/dashboard/src/elements/`. ### server Contains the main application code for the Gram
tawanorg/claude-sync/CLAUDE.md
prompt (retry passphrase / clear remote / abort). ### MCP sync `~/.claude.json` holds global MCP server configs. Unlike regular sync, MCP uses a **three-way merge** (`internal/sync/mcp.go`) against a baseline stored in `SyncState.MCPBaseline
8b-is/smart-tree/CLAUDE.md
optimized, quantum, markdown, JSON, CSV) | | MCP | `mcp.rs`, `src/mcp/` | Model Context Protocol integration (30+ tools, stdio/HTTP) | | Daemon | `daemon.rs`, `daemon_client.rs`, `daemon_cli.rs` | HTTP API server for AI context | | Proxy | `src/proxy/` | LLM proxy with
avivl/claude-007-agents/CLAUDE.md
GitHub MCP, Task Master MCP, Context7 MCP, Basic Memory MCP, Sequential Thinking MCP, Zen MCP Server **Quality**: Trunk.io linting, Pyright type checking (Python), Cockatiel/Hyx/GoBreaker resilience, Pino/Zap logging ### Agents by Category
blacklanternsecurity/red-run/CLAUDE.md
sanitized output | | `/red-run-train` | Planned | Training mode — guided walkthrough with explanations | All variants share state.db, MCP servers, and technique skills. An engagement started with one can be resumed with another. Invoke
prassanna-ravishankar/repowire/CLAUDE.md
MCP stdio to daemon HTTP MCP - `daemon-go/service/agent_backends.go` — per-backend command, capability, and resume configuration - `daemon-go/relayserver/` + `daemon-go/relay/client.go` — hosted relay server plus daemon relay client - `daemon
SepineTam/mcp-for-stata/CLAUDE.md
complete file listings from the codebase. ## Project Stata-MCP is a Python 3.11+ MCP server and CLI for running Stata do-files and inspecting statistical data. It uses FastMCP
IvanMurzak/Godot-MCP/CLAUDE.md
Godot analog of `unity-mcp-cli`), and `Godot-MCP.Tests/` (the CI xUnit suite). The MCP server is NOT in this repo — the addon consumes the shared [GameDev-MCP-Server](https://github.com
ory/lumen/CLAUDE.md
package metadata ├── scripts/ # Shared launchers and platform wrappers ├── cmd/ │ ├── root.go # Cobra root command │ ├── stdio.go # MCP server │ ├── hook.go # Hook handlers │ ├── clean.go # Age-based index data cleanup │ └── index.go # CLI indexing ├── internal/ │ ├── config
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.