GitHub 日报 · 2026 年 6 月 5 日:Anthropic 自家安全技能挤进前列

本期增速榜

1. 1
   pewdiepie-archdaemon/odysseus Python

   Self-hosted AI workspace.

   +142
2. 2
   anthropics/defending-code-reference-harness Python

   Skills for threat modeling, scanning, triage, patching, plus an autonomous scanning harness you can /customize

   +61
3. 3
   NousResearch/hermes-agent Python

   The agent that grows with you

   +60
4. 4
   colbymchenry/codegraph TypeScript

   Pre-indexed code knowledge graph for Claude Code, Codex, Gemini, Cursor, OpenCode, AntiGravity, Kiro, and Hermes Agent — fewer tokens, fewer tool calls, 100% local

   +52
5. 5
   alibaba/open-code-review Go

   Open-source & free — Battle-tested at Alibaba's scale. Hybrid architecture code review tool: deterministic pipelines + LLM Agent, precise line-level comments, built-in fine-tuned ruleset (NPE, thread-safety, XSS, SQL injection), OpenAI & Anthropic compatible.

   +46
6. 6
   chopratejas/headroom Python

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

   +44
7. 7
   affaan-m/ECC JavaScript

   The agent harness performance optimization system. Skills, instincts, memory, security, and research-first development for Claude Code, Codex, Opencode, Cursor and beyond.

   +40
8. 8
   obra/superpowers Shell

   An agentic skills framework & software development methodology that works.

   +36
9. 9
   lfnovo/open-notebook TypeScript

   An Open Source implementation of Notebook LM with more flexibility and features

   +35
10. 10
    multica-ai/andrej-karpathy-skills

    A single CLAUDE.md file to improve Claude Code behavior, derived from Andrej Karpathy's observations on LLM coding pitfalls.

    +35