Security guardrails for Claude Code, MCP tools, and Claude cowork workflows. Local-first modular YARA-style guard packs for secrets, exfiltration, prompt injection, MCP abuse, and risky agent actions.
A self-healing scraper for hostile sites: broken selectors repair themselves, browser rendering kicks in when needed, and a coherent identity layer (Chrome TLS fingerprints, matched personas, vendor-aware Cloudflare/Akamai/DataDome handling) works to slip past bot detection. Ships with an MCP server so any LLM can drive a full crawl by conversation.
Deconstructing Claude Code — interactive source code analysis, guided wc/helpe walkthroughs, and deep architectural exploration of Anthropic's AI coding agent.
Transparent proxy for Claude Code that auto-routes image-bearing requests to a multimodal model — so a non-multimodal primary model never crashes your long-running sessions.
Local Python CLI for reset credits, rate-limit windows, local usage metadata, read-only online usage/profile data, and optional API organisation usage.