Discovery
All entries

Tag

AI agent security

44 entries tagged with #agent-security.

See the curated guide →

Specifically agent-focused security: execution sandboxes, MCP scanners, LLM-as-a-judge proxies, and governance toolkits.

GitHubHackFeatured

Claude Code Analysis - architectural reverse-engineering of the leaked source

82 docs and 15 diagrams mapping every major subsystem of Claude Code's accidentally exposed 512K-line TypeScript source - YOLO classifier, 93% context compaction, prompt-cache layout, 88+ feature flags, the custom React-Fiber terminal renderer.

Why I saved this - Useful primary source for anyone building a coding agent - the YOLO two-stage classifier, the cache-busting after MCP instructions, and the 6 compaction strategies are the bits nobody else has documented.

Browse other tags