Content
85%Weight 40%Scale 1-3Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.
A well-organized, highly actionable spec-only skill body with a clear sequenced workflow, validation checkpoints, and properly signaled references. Its main weakness is token redundancy: the claims-writing procedure and automation red line are each repeated across multiple sections.
Suggestions
State the claim-writing procedure (claims.ndjson via an authorized operation: propose to registry-events.py) once in Instruction step 6 and reference it from Skill Contract → Writes and Save Results instead of repeating the full sentence three times.
Consolidate the automation red line (no ffmpeg/TTS/upload; 抖音/视频号/小红书 automation = 风控红线) into the single Scope guard block rather than restating it in the intro paragraph and the Data Sources table.
Lead the opening paragraph with the concrete beat-sheet deliverable and defer the ECHO-loop / C10 / C2 framework mapping to echo-benchmark.md, which is already referenced, to cut jargon tokens Claude does not need up front.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The body is mostly efficient and does not explain concepts Claude already knows, but the same claim-writing procedure (memory/events/claims.ndjson via an authorized operation: propose to registry-events.py) is stated in three places (Skill Contract → Writes, Instruction step 6, and Save Results) and the automation red line is repeated in the scope guard and the Data Sources table, so it could be tightened rather than earning the lean score-3. | 2 / 3 |
Actionability | Guidance is concrete and specific — exact beat timings (0-2s hook / 2-5s confirmation / 5-15s payoff / loop-or-CTA), a named list of hook families, precise read/write memory paths, paste-ready Quick Start prompts, and explicit [needs source] / Estimated labeling rules — which clears the actionable bar for an instruction-only skill even without code. | 3 / 3 |
Workflow Clarity | The 8 numbered steps are clearly sequenced with explicit validation checkpoints (step 1 NEEDS_INPUT / ECHO E1 gate, step 6 disclosure-and-claims pass with a decision rule, step 7 slop self-check) and an explicit 'Done when' contract, matching the score-3 anchor with feedback loops rather than the score-2 implicit-checkpoint case. | 3 / 3 |
Progressive Disclosure | The body is an overview with clearly signaled one-level-deep references in a dedicated Reference Materials section (echo-benchmark.md, platform norm cards, skill-contract.md, SECURITY.md, humanizer-slop.md) each with a one-line description, and the inline 8-step process is appropriately kept in SKILL.md rather than deferred. | 3 / 3 |
Total | 11 / 12 Passed |