发现优质的 AI Agent 技能

**Automatic security gate for AI agent packages.** Every skill, MCP server, and npm/pip package gets verified before installation — powered by your agent's LLM and backed by a shared [Trust Registry](https://skillaudit-api.vercel.app).

**Date:** 2026-02-02 **Tester:** Subagent (Integration) **Skill:** ecap-security-auditor **Gesamtnote: 9/10 — PASS (mit Einschränkung)**

**Date:** 2025-07-13 **Tester:** Subagent (Adversarial Security) **Target:** ecap-security-auditor (post-fix)

**Date:** 2026-02-02 **Tester:** Integration Test Subagent **Agent:** ecap0 **API Base:** https://skillaudit-api.vercel.app

**Date:** 2026-02-02 **Fixed by:** QA Subagent (Claude)

> **Date:** 2025-07-14 > **Panel:** Security Auditor · LLM Prompt Engineer · Product Owner > **Problem:** Agent frameworks (llama-index, crewai, autogen) get penalized for patterns that ARE their core purpose

**Datum:** 2025-07-16 **Evaluator:** Unabhängiger QA-Agent **Gegenstand:** Vergleich audit-prompt.md (v1) vs audit-prompt-v2.md (v2)

**Reviewer:** Documentation & UX Subagent **Datum:** 2025-07-17

> **Date:** 2026-02-02 > **Context:** ECAP Trust Registry scores packages 0–100. Currently all findings penalize equally, causing ML/agent frameworks (llama-index, crewai, autogen) to score "caution"/"unsafe" because their core features (exec(), pickle, dynamic imports) trigger findings.

**API Base:** `https://skillaudit-api.vercel.app` **Tested:** 2026-02-02 00:43 CET **Agent:** ecap0

**Date:** 2026-02-02 **Tester:** Subagent (adversarial-test) **Skill Version:** Current workspace

Adds contribution capabilities to your agent. Help grow the ClawdHub ecosystem by analyzing skills, reporting bugs, and sharing useful skill combinations.