Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →ibrahim-3d/loop-execution-evaluator
Evaluate-Loop Step 4: EVALUATE EXECUTION. This is the dispatcher agent — it determines the track type and invokes the correct specialized evaluator. Does NOT run a generic checklist. Instead dispatches to: eval-ui-ux (screens/design), eval-code-quality (features/infrastructure), eval-integration (APIs/auth/payments), eval-business-logic (generator/rules/state). Triggered by: 'evaluate execution', 'review implementation', 'check build', '/phase-review'. Always runs after loop-executor.
Passed 13 of 17 test cases with the skill, and 5 of 17 without it.
and 10 more in the full report
10 fixed7 passed either way5 still failing0 broke
Tokens, turns and time describe resource use, not a dollar price.
One of 10 test cases the skill fixed. Same model, same prompt, with and without the skill.
During Step 4 structural verification, what shell build command is routinely verified for a zero exit status across JavaScript project tracks?
| Checked | Without skill | With skill |
|---|---|---|
| Executes npm run build to verify standard project build status | Missed | Passed |
The instructions your model reads when the skill loads.
This agent does NOT evaluate directly. It determines the track type and dispatches the correct specialized evaluator.
Different track types need fundamentally different checks:
A generic checklist misses critical issues specific to each type.
read_file the track's metadata.json and spec.md to determine the track type, then dispatch:
| Track Type | Keywords in spec/metadata | Evaluator |
|---|---|---|
| UI / Design | "screen", "component", "design system", "layout", "visual", "UI shell" | eval-ui-ux |
| Feature / Code | "implement", "feature", "refactor", "infrastructure", "hook", "store" | eval-code-quality |
| Integration | "Supabase", "Stripe", "Gemini", "API", "auth", "database", "webhook" | eval-integration |
| Business Logic | "generation", "lock", "dependency", "pricing", "tier", "pipeline", "download" | eval-business-logic |
Some tracks need multiple evaluators. For example:
eval-business-logic + eval-code-qualityeval-integration + eval-code-qualityeval-ui-ux onlyWhen multiple evaluators apply, run them all. The track passes only if ALL evaluators pass.
1. read_file track metadata.json + spec.md
2. Determine track type(s)
3. Dispatch evaluator(s):
test cases passed, vs 5 / 17 without
More skills from the same source and nearby in the registry, with their headline benchmark lift.