Tork Governance

Tork Governance is a productivity tool that provides a universal governance layer for AI agents, enforcing policies, blocking dangerous tools, detecting PII, and requiring human approval across frameworks including LangChain, CrewAI, AutoGen, and OpenAI MCP. The platform sits between AI agents and the outside world, governing every tool call, API request, and database query before execution. Tork supports purpose-built Model Context Protocol governance, kill switches for instant agent halting, JSON-based policy rules, and PII detection for over 50 types across 13 countries running locally on user infrastructure. The free tier includes 5,000 API calls per month, while the starter plan at $29 per month offers 50,000 calls with advanced PII detection and email support. Tork provides 11 SDKs and 124 integrations including Slack, email, and dashboard for human-in-the-loop approval workflows. Tork Governance targets organizations deploying AI agents at scale who need audit trails, policy enforcement, and safety controls across their agent fleet.