Conduct - Open-source guardrails for LLM and MCP tool calls

Show HN: Conduct, open-source guardrails for LLM and MCP tool calls

Conduct is an open-source runtime governance platform for AI agents, combining a policy engine (Conduct Guard) and an LLM proxy (Conduct Router) to enforce one policy across every LLM call, shell tool, and MCP invocation. It decides block, warn, audit, or inject before actions execute, backed by signed configuration and a hash-chained audit log. With 20+ compliance packs, 22 pre-built playbooks, a canvas UI, and a CLI, Conduct ensures fail-closed security and auditability. Start with free Discovery mode to observe AI actions for 14 days, then promote rules to enforcement. Self-host via Docker or use the hosted service.