Skip to main content

2 docs tagged with "safety"

View all tags

Tool Contracts for AI Agents

Design tools whose inputs, authority, side effects, failures, and evidence remain legible to models and operators.