李韭二

api-relay-audit. Local security audit for AI API relays and LLM proxies: detects prompt injection, model substitution, tool-call rewriting, SSE anomalies, error leakage, and Web3 wallet risks.

epro-memory. LLM-powered agent memory with 6-category classification and L0/L1/L2 tiered structure

memx-memory. Persistent, research-backed user memory for AI assistants.

community-codex-windows-troubleshooting. Community field guide and read-only diagnostics for troubleshooting Codex on Windows.

github-lab-monitor. Monitor top AI research labs on GitHub using XCrawl + LLM analysis. Scrape → Structure → Analyze → Publish.

wulin-town. TypeScript