Explain PSS suggestion confidence levels and evidence types for reindexing decisions.
npx skills add https://github.com/Emasoft/perfect-skill-suggester --skill pss-usage
Please help me install this Agent Skill. Skill: pss-usage Source: https://github.com/Emasoft/perfect-skill-suggester/tree/main/skills/pss-usage Command: npx skills add https://github.com/Emasoft/perfect-skill-suggester --skill pss-usage
Interprets Perfect Skill Suggester (PSS) outputs so you can understand confidence levels and the supporting evidence before acting.
Type a plain-language prompt such as "Explain the latest PSS suggestions" to get guidance on interpreting results.
High-intent search queries and answers about installing and using this skill.
To interpret Perfect Skill Suggester outputs, check HIGH, MEDIUM, or LOW confidence signals and review the supporting evidence types like intent, keyword, and co_usage before taking activation steps.
Evidence types in PSS suggestions include intent, keyword, and co_usage, which collectively determine the confidence level of a skill recommendation and indicate why a specific suggestion was generated.
To troubleshoot empty or misaligned skill suggestions, map the PSS diagnostic outputs to concrete steps such as running status checks, performing an indexing refresh, and verifying your agent configuration.
You should reindex skills when suggestions become inaccurate or misaligned, using the reindexing workflow to refresh the indexing status and ensure your skill activation decisions are based on current data.
Map diagnostic concepts from the suggestion table to actionable steps by reading confidence levels and evidence, running status checks, and applying this guidance to determine which skills to activate.
Yes, running status checks before skill activation is recommended to verify the indexing state and ensure the PSS suggestion table is accurate and actionable for your agent configuration.
grtninjaConfigure REPO_B Agent Bridge read-only mode and gate write operations.
EmasoftValidate skill directories and trigger PSS reindexing for AMCOS governance.
dereekbProvide foundational Angular utilities for routing, storage, context, and authentication.
ken-cavanagh-gleanEvaluate LLM outputs with direct scoring, pairwise comparison, and rubric generation.
