Skill Review Protocol • AI Agent SDK
SkillReviewProtocol: Policy for the autonomous skill self-improvement loop.
SkillReviewProtocol
Defined in the protocols module.
AI Agent
Policy for the autonomous skill self-improvement loop.
After a task finishes, an opt-in guarded review pass asks the agent whether the session revealed a reusable technique worth capturing as a skill. `should_review decides whether to run the pass and review_prompt` produces the directive given to the agent. Both take a lightweight trajectory dict so wrappers/plugins can swap the policy without depending on core internals.
Methods
should_review()
Return True if the finished session warrants a skill review.
review_prompt()Return the directive prompt for the guarded review turn.
Source
View on GitHubpraisonaiagents/skills/protocols.py at line 201