skill-assessor

Auto-trigger eligibility depends on systematic evaluation of safety guardrails, frequency of use, convenience, and token cost. This Skill assesses those criteria across your skill repository to recommend whether a skill should auto-invoke or remain manual-only, preventing skill bloat and optimizing performance.

Assessment prioritizes safety guardrails first, then frequency of applicable scenarios, convenience gains, and mandatory token-cost analysis. Multi-phase evaluation orchestrates context gathering and code-pattern analysis to generate structured recommendations with trigger configurations and implementation steps.

Yes. The Skill includes token-cost analysis as a core requirement, evaluating whether auto-invocation provides sufficient value relative to computational expense. This ensures your skill repository stays cost-efficient while maintaining high-frequency triggering only where justified.

Beyond recommending AUTO-INVOKE or MANUAL-ONLY, the Skill generates trigger configuration options, detailed rationale grounded in code analysis and domain context, and implementation steps to deploy your recommendation across the skill repository.

Safety guardrails are the prioritized criterion during evaluation. The Skill analyzes whether new skills align with safety requirements before considering frequency or convenience, ensuring guardrail-critical skills trigger appropriately across all scenarios.

Provide the skill name and its repository location. The assessor orchestrates deep context gathering, code-pattern analysis, and optional domain research to conduct comprehensive multi-phase evaluation without requiring manual input beyond the initial skill identifier.