Code-only MCP servers — agent-tools.cloud

MCP servers that ship as code — a repository or a package you install and run locally (usually over stdio). They have no hosted endpoint, so we cannot probe or grade them: instead each one shows its repository, install command, license, stars and when it was last updated.

Collected from GitHub repositories and npm, PyPI, Cargo, Go, Maven, NuGet and RubyGems packages tagged as MCP servers, the official MCP Registry, Docker's MCP catalog, ModelScope, Baidu MCP World, Cline and other public directories; one entry per repository. A server that also offers a hosted endpoint is listed under Remote instead. These entries are not counted in our reachability, verification or readiness statistics.

tokenburnmarket

github.com/formidable-oss/tokenburnmarket
active

Connect your machine to tokenburnmarket: sync agent usage via ccusage, run the MCP server, place bets.

npm MIT
ccenv

github.com/sshoecraft/ccenv
active

Self-contained Claude Code env: bundled MCP servers (ccmemory, ccusage, ccteam), project awareness skill (ccproject), and a relay-loop wrapper (ccloop), with an overlay system for per-user / per-host extensions.

MIT
viberank-mcp

github.com/sculptdotfun/viberank
active

MCP server for viberank: submit your AI coding usage (Claude Code, Codex, Gemini CLI and more) and read the leaderboard

npm MIT
tokenarchive

github.com/ddyy/tokenarchive
active

Daily snapshots of your AI coding usage, kept forever: local dashboard, optional Cloudflare Worker backup, MCP servers over both.

MIT
claude-usage-panel

github.com/fschmutz/claude-usage-panel
active

See your Claude Code plan usage everywhere: GNOME top bar, macOS menu bar, a status line under the Claude Code prompt, and a get_usage MCP tool for Claude Code + Cursor. Session, weekly & per-model (Fable/Opus) limits from the official usage API, with burn-rate forecasts. One-line install, self-updating.

MIT ★ 8