724
total weekly downloads
4 packages
logbook-agent
Local-first, tamper-evident audit trail for AI agents. The operator holds the only key; the agent vendor can neither read nor forge the record. Includes an MCP proxy that records what an agent actually did, not what it self-reports.
toolproof-lock
Versioned, human-approved baselines for agent capabilities — fail CI when an MCP/API capability surface changes
toolproof-mcp
Guard rail for the AI tool supply chain — wrap any MCP server to strip malicious tools before your agent sees them, redact injections from live tool results, grade every API, and fail CI when a tool turns hostile.
toolproof-scan
Trust grades for MCP servers and APIs — the verify step for the agent economy