Content
67%Scale 1-5Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.
This is a well-structured, concise skill that clearly defines a discovery/gathering workflow with good decision logic for how to fill information gaps. Its main weakness is limited actionability — it tells Claude what kinds of things to do but lacks concrete examples of questions to ask, specific hunting commands, or a checklist of what 'enough context' looks like. The referenced bundle file for hunting artifacts would likely address some of this gap.
Suggestions
Add 1-2 concrete example probe questions for common scenarios (e.g., 'What triggers this skill?' or 'Walk me through the last time you did this manually') to improve actionability.
Define explicit criteria for 'enough' in section 4 — e.g., a short checklist: problem statement ✓, trigger condition ✓, key steps ✓, constraints ✓ — to strengthen the validation checkpoint.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The content is efficient and respects Claude's intelligence. It avoids explaining basic concepts and uses terse, structured guidance. Minor verbosity in section 2's bullet explanations (e.g., 'This is often the starting point') could be trimmed, but overall it's lean. | 4 / 5 |
Actionability | The skill provides concrete guidance on what to look for and how to fill gaps (ask vs hunt vs infer), with specific file paths like `.tessl/memory/preferences/`. However, it lacks executable examples — no sample questions to ask, no concrete commands for hunting artifacts, and the probe example is generic. It's instructional rather than executable. | 3 / 5 |
Workflow Clarity | The four-step sequence (understand → stock → fill gaps → confirm) is clear and logical with a well-defined stopping condition. The decision tree for filling gaps (ask/hunt/infer) adds useful structure. Minor gap: no explicit validation checkpoint — 'confirm you have enough' is stated but lacks concrete criteria for what 'enough' looks like. | 4 / 5 |
Progressive Disclosure | The skill appropriately references `references/hunting-artifacts.md` for detailed hunting guidance rather than inlining it, and points to preference files for grounding. However, the bundle has no files provided, so we can't verify the referenced path exists. The structure is clean with well-signaled sections, though the single reference is the only disclosure point. | 4 / 5 |
Total | 15 / 20 Passed |