MCP servers and skills: search, filter by where they run, and compare.
MCP server that runs shell commands through RTK (Rust Token Killer), so Claude Desktop and other MCP clients get compact command output. Rust.
SkillMITFew starsNo key
A Model Context Protocol (MCP) server for enhanced shell command execution — RTK token minimization, persistent caching, execution logging, and safe file writes for AI coding agents
Node.jsMIT★ 2Few starsNo key
Token-optimizing MCP proxy - sits between Claude and any MCP server, compressing tool responses by 60-90% through an 8-stage filter pipeline.
SkillMIT★ 4Few starsNo key
Model Context Protocol (MCP) Server, Subagent Orchestrator & Lifecycle Governor for Google Antigravity (Powered by RTK)
Node.jsMITFew starsNo key
Codex Agent Skill for token optimization with RTK (Rust Token Killer) and Headroom MCP: compress shell outputs, large tool outputs, logs, files, and report measured token savings.
SkillMIT★ 3Few starsNo key
Orchestration + observability + coach layer for Claude Code token optimization. 13 MCP tools, 238 tests, measurement honesty.
Node.jsMIT★ 3Few starsNo key
One memory, every coding agent — shared brain + usage dashboard for Claude Code, Codex, Gemini (subscription plans, no API keys)
SkillMIT★ 3Few starsNo key
Lightweight, evidence-preserving coding workflows for Claude Code, Codex, OMP and Pi.
SkillMIT★ 25No key
⚡ Claude Adaptive Token Optimizer — Zero-dependency, cache-preserving token optimization toolkit for Claude Code CLI and Desktop
PythonMIT★ 1Few starsNo key
Cut token cost in 4 layers, without cutting model intelligence: handoff, rtk, headroom, caveman — token-saving skills for DeepSeek Harness.
SkillMIT★ 2Few starsNo key
One-command setup that makes Claude Code use fewer tokens: rtk shell compression, stash search index, web-search, codebase-memory, fast-jev-compaction, ponytail, plus Jev skill suggestions and model r
SkillMIT★ 1Few starsNo key
Agent skill that routes each coding task to the most token-efficient tool per layer: serena for code reads, rtk for command output, caveman for prose, Ponytail for generated code — −70% tokens measure
SkillMIT★ 12No key
Token-saving starter pack for Claude Code, Codex and other coding agents — paste one line, the agent installs the lot.
SkillMITFew starsNo key
Never stop coding. Free AI gateway: one endpoint, 160+ providers, RTK+Caveman stacked compression up to ~95% eligible context savings, smart auto-fallback, MCP/A2A, multimodal APIs, Desktop/PWA.
Node.jsMIT★ 3Few starsNo key
Self-hosted recorder, map and guarded control panel for Yarbo robot mowers. Independent community project, not affiliated with Yarbo.
SkillMIT★ 8No key
Token optimization layer for Claude Code: installs and wires RTK, tilth, Serena, and ICM so your AI agent reads what it needs, not everything.
SkillMIT★ 1Few starsNo key
concierge is a set of skills for claude-code that allows a very opinionated usage of gastown (and some other tools) to speed even further creation/maintenance of projects.
SkillMIT★ 3Few starsNo key
Cross-host operational companion for Compound Engineering across AI coding agents. Fleet version governance, project adoption, agent guardrails, and advisory workflow observation.
SkillMIT★ 6No key
Query-aware context compression for LLMs and coding agents. Cut ~65% of input tokens.
Node.jsMIT★ 138No key
One-command bootstrap for a fully wired Claude Code workstation — installs Claude Code, RTK, context-mode, code-review-graph, caveman, hooks, MCP servers, statusline, and 39+ curated skills. macOS + L
SkillMIT★ 8No key
Opencode plugin for Entire CLI for Git Hook AI Session capture
SkillMIT★ 1Few starsNo key
Tiny Python client for the open KUON GEO API (kuongeo.com)
Hosted by authorMITFew starsNo key
Who stands behind it
