SKILLS & AGENTS/Skill 与 Agent·MCP-SERVER

headroom

headroomlabs-ai/headroom

Compress tool outputs, logs, files, and RAG chunks before they reach the LLM. 20% fewer tokens for coding agents, 60-95% fewer tokens for JSON, same answers. Library, proxy, MCP server.

打开仓库Open repository 官网Homepage

怎么装：Install: uv tool install --python 3.13 "headroom-ai[all]" # CLI in a self-contained env

02 / 现在的位置02 / Why now

它现在处在什么位置Where it stands right now

首次发现FIRST SEEN

03 OCT 202603 OCT 2026

增长GROWTH

+0.1%+0.1%

74k → 74k stars · 2 个快照74k → 74k stars · 2 snapshots

状态STATUS

持续活跃Active

03 / 它能帮你做什么03 / What it helps you do

它能帮你做成什么What it helps you do

04 / 安装04 / Install

怎么装How to install

$

装在独立环境里，不污染系统 Python。带完整 CLI。Installs into its own isolated environment without touching your system Python. Ships the full CLI.

NPM

$

只装 TypeScript SDK，没有命令行工具，适合写进自己的项目里调用。Installs only the TypeScript SDK, with no CLI — for calling from your own project.

两种装法选一个即可 —— 上面那条给命令行用户，下面那条给写代码的用户。Pick one of the two — the first is for command-line users, the second for people writing code.

05 / 限制与风险05 / Limits and risk

限制与风险Limits and risk

风险不是警告，是可信度的一部分。以下结论只基于文档静态扫描，我们不会执行项目里的任何代码。Risk here is evidence, not an alarm. These findings come from static scanning of the docs; we never execute a project’s code.

限制LIMITS

存在风险Risk found

01需要 API KeyREQUIRES API KEY: 是Yes
02需要本地环境NEEDS LOCAL RUNTIME: 是Yes
03需要 DockerNEEDS DOCKER: 未在文档中发现Not found in the docs
04难度DIFFICULTY: 需要本地环境Needs a local runtime

PROVENANCE · 这些结论的依据是什么Where these conclusions come from

来源 URLSource URL: https://github.com/headroomlabs-ai/headroom
原始数据哈希Raw data hash: 72fae8c74209e4b3b4a14e5acafc5dbd
解析时间Parsed at: 2026-10-04T09:07:05.197Z
最后一次验证Last verified: 2026-10-04T08:53:33.891Z
增长测量时间Growth measured: —
发现查询Discovery queries: focus:token:repo:topic-context-engineering, focus:token:repo:topic-token-optimization
分类来源Category source: rules
焦点Focus: token (17)

静态扫描STATIC SCAN

01执行远程脚本Runs a remote scriptreadme

curl --proto '=https' --tlsv1.2 -sSf https://sh.rustup.rs | sh && rustup default stable

02读取 API KeyReads an API keyreadme

**Platform support.** macOS auth reuse through Copilot CLI Keychain storage and

这里只做静态扫描：读 README、SKILL.md 和依赖清单，不执行代码。没有命中不代表安全。This is static scanning only: we read the README, SKILL.md and dependency list, and never execute code. No findings does not mean safe.

完整解析证据 JSONFull parse evidence JSON

{
  "agents": [
    {
      "agent": "claude-code",
      "evidence": "tree: .claude/"
    },
    {
      "agent": "cursor",
      "evidence": "readme: …roxy and prints base URLs for Cursor settings | | Aider | ✅ | starts proxy…"
    },
    {
      "agent": "cline",
      "evidence": "readme: …copilot|cursor|aider|opencode|cline|continue|goose|openhands|open…"
    },
    {
      "agent": "vscode",
      "evidence": "readme: …| starts proxy + launches | | VS Code Copilot | ✅ | transparent pro…"
    },
    {
      "agent": "opencode",
      "evidence": "readme: …dex|grok|copilot|cursor|aider|opencode|cline|continue|goose|openhand…"
    },
    {
      "agent": "copilot",
      "evidence": "readme: …path. <details> <summary><b>GitHub Copilot CLI subscription mode</b></su…"
    }
  ],
  "apiKey": "ker and CI, pass an explicit `GITHUB_COPILOT_TOKEN` or `GITHUB_COPILOT_GITHUB_TOKEN` inste",
  "docker": null,
  "taxonomy": {
    "scores": {
      "design": 0,
      "ai-tools": 11,
      "dev-tools": 0,
      "indie-web": 0,
      "skill-agent": 9,
      "productivity": 9
    },
    "primary": "skill-agent",
    "secondary": "mcp-server"
  },
  "localRuntime": "─────────┐ │ Headroom (runs locally — your data stays here) │ │ ─────",
  "skillMdTotal": 0,
  "skillMdErrors": [],
  "treeTruncated": false,
  "categoryScores": {
    "data": 3,
    "agent": 0,
    "media": 0,
    "design": 0,
    "browser": 0,
    "devtool": 0,
    "security": 0,
    "marketing": 0,
    "productivity": 0
  },
  "classification": [
    {
      "signal": "server.json (MCP registry manifest)",
      "weight": 0.85
    },
    {
      "signal": "python dependency on mcp / fastmcp + MCP framing",
      "weight": 0.8
    },
    {
      "signal": "Claude Code plugin manifest",
      "weight": 0.7
    },
    {
      "signal": "MCP topic + README",
      "weight": 0.6
    },
    {
      "signal": "description mentions AI",
      "weight": 0.2
    }
  ],
  "scannedSources": [
    "readme"
  ],
  "skillMdFetched": 0
}