Convert raw agent features into testable Skills with SKILL.md contracts and resolver routing.
npx skills add https://github.com/weiping/gbrain-cn --skill skillify-weiping
Please help me install this Agent Skill. Skill: skillify Source: https://github.com/weiping/gbrain-cn/tree/main/skills/skillify Command: npx skills add https://github.com/weiping/gbrain-cn --skill skillify-weiping
Skillify converts a raw feature or workflow into a discoverable, reliably routed, and quality-checked Skill with tests and documented operational behavior.
SKILL.md, deterministic code where applicable, eval receipts, tests, resolver routing, and optional brain filing.Use the skillify workflow to skillify your feature by creating skills/<slug>/SKILL.md, extracting deterministic logic to scripts/, running cross-modal eval, and then adding resolver and tests so the behavior is consistently reachable and verifiable.
High-intent search queries and answers about installing and using this skill.
Skillify converts raw agent features into testable capabilities by extracting deterministic logic to scripts, defining a SKILL.md contract, and adding trigger phrases for resolver routing to ensure consistent invocation.
Cross-modal evaluation is a quality gate that runs three frontier models from different providers to score and iterate on a documented dimension list, ensuring robust agent behavior before tests lock in the logic.
You add trigger phrases to resolver routing during the skillify workflow, then verify correct routing via resolver eval and perform orphan, overlap, and DRY audits to ensure safe and reachable behavior.
Yes, a SKILL.md contract is required to define operational phases, document behavior, and pass the completeness checklist for unit, integration, and LLM evaluations before achieving operational readiness.
The best way to verify agent capabilities end-to-end is to validate trigger-to-side-effect behavior, produce durable eval receipts, and run resolver checks alongside DRY and orphan audits for operational readiness.
inbrainfunGenerate SKILL.md, test suites, and resolver entries from a raw feature.
mr-daedaliumMerge PRs, trigger deployment, and verify canary health in CI/CD pipelines.
latestaiagentsTest AI agent systems from unit to end-to-end with mock LLM responses.
starlink-awakenStandardize agent skill development through multi-phase audits and cross-modal evaluation.