Content
100%Weight 40%Scale 1-5Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.
An exemplary concise skill body: exact action syntax, concrete input examples, an explicit feedback loop to a real one-level-deep reference file, and a clear success criterion — all in about 20 lines with no padding. Nothing material is missing for the task it covers.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The ~20-line body is lean and fully operational — command template, inputs with examples, three process steps, and expected output — with no explanation of concepts the agent already knows. Every token earns its place, matching the top anchor. | 5 / 5 |
Actionability | It provides the exact executable action format "take {object} from {source_receptacle}" plus concrete identifier examples ("toiletpaper 1", "toilet 1") and the exact failure signal to watch for ("Nothing happened"). For an instruction-only skill this is copy-paste-ready guidance covering the common case. | 5 / 5 |
Workflow Clarity | The simple-skill exception applies: a single unambiguous action with a clear three-step sequence (verify context, execute, handle feedback), an explicit error-recovery loop routing "Nothing happened" to the troubleshooting guide, and a defined success observation in the Output section. Not a destructive or batch operation, so no cap applies. | 5 / 5 |
Progressive Disclosure | The body is under 50 lines, well-organized into Inputs/Process/Output, and its single reference (references/troubleshooting.md) exists in the bundle, is one level deep, and is signaled exactly at the point of need in the Handle Feedback step. This matches both the top anchor and the simple-skill scoring note. | 5 / 5 |
Total | 20 / 20 Passed |