[ { "query": "run the evals for my brief-writer skill", "should_trigger": true }, { "query": "does bmad-ticket still beat the bare model? run a baseline", "should_trigger": true }, { "query": "benchmark the skill with 5 runs per case so I get a stable mean", "should_trigger": true }, { "query": "check whether the description of skills/deploy-helper fires on the right requests", "should_trigger": true }, { "query": "grade the outputs in evals/cases.json against their rubrics", "should_trigger": true }, { "query": "compare the stripped version of my skill against the full one on the same cases", "should_trigger": true }, { "query": "my skill keeps loading when people ask about unrelated stuff, can you measure and fix the triggers", "should_trigger": true }, { "query": "I have an evals.json from skill-creator, run those cases through the skill", "should_trigger": true }, { "query": "evaluate the research-brief skill before I ship it", "should_trigger": true }, { "query": "is section 3 of my SKILL.md doing anything? test with and without it", "should_trigger": true }, { "query": "write a skill that drafts release notes", "should_trigger": false }, { "query": "run the tests", "should_trigger": false }, { "query": "evaluate this PR and tell me if it is safe to merge", "should_trigger": false }, { "query": "benchmark the API with 1000 requests per second", "should_trigger": false }, { "query": "grade my essay on the French Revolution", "should_trigger": false }, { "query": "compare the two versions of this React component and pick the cleaner one", "should_trigger": false }, { "query": "fix the typo in the description of my deploy-helper skill", "should_trigger": false }, { "query": "evaluate whether we should migrate from Jest to Vitest", "should_trigger": false }, { "query": "pytest is failing on the trigger module, help me debug it", "should_trigger": false }, { "query": "write a rubric for grading student code submissions", "should_trigger": false } ]