GStars
    headroomlabs-ai

    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.

    ai-agents
    ai
    backend
    llm
    networking
    agent
    anthropic
    claude-code
    compression
    context-engineering
    Python
    Apache-2.0
    78.8K stars
    6.2K forks
    78.8K GitHub watchers
    Updated Nov 4, 2026
    View on GitHub

    Build with Backblaze B2

    SDKs, agent skills, IDE extensions, and reference pipelines from Backblaze Labs. All open source.

    Explore Backblaze Labs

    Loading star history...

    Where it stands

    Rank, growth pace and release cadence relative to everything GStars tracks.

    5 releases in the last 90 days, the latest 8 days ago, with 355 contributors on GitHub.

    Peaked at #58 in the weekly ranking the week of Apr 5, its best in the past year.

    Use Cases & Benefits

    • Compresses AI agent inputs like tool outputs, logs, files, and RAG chunks to reduce token usage before sending to large language models.
    • Provides lossless, multi-algorithm compression with reversible context retrieval, preserving accuracy while significantly reducing token counts.
    • Use for optimizing token consumption in coding agents interacting with models like Claude Code, Codex, or Cursor to lower API costs.
    • Use for compressing JSON, code, and conversation history in multi-agent workflows to share context efficiently across different AI agents.
    • Use for integrating transparent proxy compression in existing Python or TypeScript LLM applications without code changes for seamless token optimization.

    Should you depend on it?

    License terms, project hygiene and risks, from GitHub metadata and GStars' snapshots.

    License

    Apache 2.0

    Permissive: commercial use, modification and redistribution are allowed with attribution; includes an explicit patent grant from contributors.

    Community files

    • README: present
    • Contributing guide: present
    • Code of conduct: present

    Watch-outs

    No watch-outs from GStars' checks.

    General guidance from repository metadata, not legal advice. How we score

    From the README

    Compress everything your AI agent reads. Same answers, fraction of the tokens.

    Headroom sits between your application and the LLM provider. It intercepts requests, compresses the context, and forwards an optimized prompt. Use it as a transparent proxy (zero code changes), a Python function (compress()), or a framework integration (LangChain, LiteLLM, Agno).

    Excerpt from the headroomlabs-ai/headroom README on GitHub (Apache-2.0 license). © its authors.Read the full README on GitHub

    Daily snapshots

    Stars gained per day, and how forks and open issues moved, from GStars' daily snapshots

    Oct 10: +36Oct 21: +512Oct 22: +500Oct 23: +304

    +12,802 stars over 85 days across 11 snapshot days; best day +512 on Oct 21.

    Forks

    6.2K+1.1K

    Open issues

    470-167

    Compared with peers in AI Agents

    Fastest-growing repositories GStars tracks in this cohort for the week of Nov 8, by 7-day star growth

    View category →
    headroomlabs-ai/headroom compared with AI Agents repositories by stars, 7-day star growth, health score and last release
    RepositoryStars+7dHealthLast release
    morluto/rea450.0K+22.9K7516 days ago
    vercel-labs/agent-desktop21.4K+8.7K755 days ago
    milind-soni/OpenMausBot20.6K+8.4K75today
    stablyai/orca121.4K+6.7K75today
    thedotmack/claude-mem115.8K+4.7K75today
    headroomlabs-ai/headroom78.8K—658 days ago
    Category median115.8K+8.4K75—
    Compare headroom with rea, agent-desktop, OpenMausBot →

    Discover Repositories

    Search across tracked repositories by name or description