Skip to content
View lennney's full-sized avatar

    Highlights

    • Pro

    Block or report lennney

    Block user

    Prevent this user from interacting with your repositories and sending you notifications. Learn more about blocking users.

    You must be logged in to block users.

    Maximum 250 characters. Please don’t include any personal information such as legal names or email addresses. Markdown is supported. This note will only be visible to you.
    Report abuse

    Contact GitHub support about this user’s behavior. Learn more about reporting abuse.

    Report abuse
    lennney/README.md

    Hi, I'm Guohao Liu 👋

    I'm an AI engineer with a product mindset. I build tools and infrastructure for AI agents, with a focus on reliability, context efficiency, search, and security.

    My work starts with a real workflow. I study it, build the smallest useful version, ship it, and use feedback and evidence to decide what comes next. I work across the full loop: product thinking, implementation, launch, and iteration.

    Right now, I'm exploring:

    • MCP and infrastructure for AI agents;
    • context efficiency, reliability, and security;
    • search and research workflows for agents;
    • evaluating AI tools with reproducible evidence.

    I also contribute to open source and write about what I build, what failed, and what I learned.

    take-a-deep-breath0.com

    Pinned Loading

    1. stop-that-shitstop-that-shitPublic

      Stop That Shit(别再造史了)|面向 Codex/GPT 场景的多平台 Hook + Skill Guard:拦截 AI coding agent 无需求的哈希、校验和与任务范围膨胀。 A multi-platform Hook + Skill Guard for AI coding agents in Codex/GPT workflows: stop unrequested …

      JavaScript 226 14

    2. BigPizzaV3/CodexPlusPlusBigPizzaV3/CodexPlusPlusPublic

      An enhanced tool for CodexApp, striving to make Codex better to use and more comfortable 一个CodexApp的增强工具,努力让Codex变得更好用更舒服

      Rust 29.4k 1.9k

    3. headroomlabs-ai/headroomheadroomlabs-ai/headroomPublic

      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.

      Python 67.2k 5.2k

    4. gate-all-skillsgate-all-skillsPublic

      Frontend/Next.js Claude Code Skills — curated + custom

      Python 1 1

    5. mcp-slim-guardmcp-slim-guardPublic

      Context compression for MCP. Same upstream call, exact results recoverable.

      TypeScript 120

    6. agent-search-mcpagent-search-mcpPublic

      Free-first Chinese and English web search MCP using zero-key sources and inspectable evidence.

      TypeScript 107 8