Forkit Dev AI Passport - AI evidence layer for models and agents

AI is moving from chat to action. Models and agents now use tools, access data, trigger workflows, and support real business decisions. But when something changes or goes wrong, teams still struggle to answer basic questions: which model or agent acted, who approved it, what changed during its lifecycle, what data or tools were accessed, and what happened at runtime? That is why we built Forkit Dev AI Passport: a living evidence record for models and agents, connecting authorship, lineage, approvals, runtime signals, deployments, and proof in one shared report. The open-source core is available on GitHub, and we would love feedback from builders, AI teams, and anyone working on production AI systems. What would you want an AI Passport to prove before trusting a model or agent in production?