See and manage everything that shapes Claude Code on your machine: instructions, settings, permissions, hooks, MCP servers, agents, skills and memory. npx ccdesk
View on GitHub ↗Homepage ↗See and change everything Claude Code loads on your machine: instruction files, settings, permissions, hooks, MCP servers, subagents, skills, plugins and auto memory.
Each piece lives in its own file, at one of six scopes: managed, user, project, local, auto memory or a plugin. ccdesk opens in your browser on the directory you ran it from, lists everything that applies there in the order Claude Code loads it, and names the file behind each piece. When something is wrong, change it on the same page: you review the diff first, and every file ccdesk writes is backed up and can be undone. It is not a session viewer or a usage dashboard.
- Why does Claude ask me before every git push here? (Permissions) - Which model does this repository use, and which file set it? (Settings) - Is my AGENTS.md loaded at all? (Instructions) - How much context does my config take before I type anything? (What Claude sees) - What do I keep approving by hand, and which allow rules have I stopped needing? (Permissions) - Which project MCP servers still wait for approval? (MCP servers)
What you see What you can change Docs --- --- --- --- Instructions Every CLAUDE.md, AGENTS.md and .claude/rules/.md that applies, in load order, with why a file is skipped and who else sees it Open it in your editor Instructions Settings The value Claude Code ends up using for each key, the file it comes from, and the values it overrides Set or remove any documented key in user, project or local settings, picked from the schema Settings Permissions Every allow, ask and deny rule, which ones never take effect, and all
From the project README.
Add the radar badge to your README — it shows your project was picked up by MCP Radar and links to this page:
[](https://mcp.liqiwa.com/s/ayagmar--ccdesk.html)
See the tokens Claude Code and Codex load before you type: MCP tool lists, skills, hooks, CLAUDE.md and AGENTS.md per session. Find what you never use and archive it, with undo. Open source, runs offl
vshulcz/deja-vuYour agents already solved this. deja finds it — retroactive search and MCP recall over the session logs your coding agents already wrote. Cross-agent, SSH-portable, one zero-dep binary, fully local.
00200200/usagetrimKeep the signal. Cut the noise. Local CLI + MCP — fold verbose tool output for Claude Code, Codex, Cursor & Desktop.
cuongnbms/mastering-claude-codeCẩm nang thực chiến Mastering Claude Code (tiếng Việt): 16 chương biên soạn từ tài liệu chính thức của Anthropic — CLI, workflows, MCP, subagents, skills, hooks, plugins, CI/CD & tối ưu context.
dontpayfull/AInotateAnnotated screenshots for AI agents: Skitch-style arrows, numbered steps, boxes, labels, solid redaction
thaildhe172591/pythiaPL/SQL development for AI agents on Oracle Database - expert queries, impact analysis, and a snapshot-verified write path with honest rollback
The top new MCP servers of the week, every Monday. No spam, unsubscribe anytime.