Content
88%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.
The body is a dense, executable runbook: concrete commands, a full output contract, explicit failure handling, and a precise downstream report/DNA format, with verified references into a small scripts bundle. Its only weaknesses are modest repetition and a couple of large inlined artifacts that would fit better in reference files.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | Nearly every section carries non-obvious operational detail (ASR provider ordering, frame-sampling percentages, exit codes, DNA template) with no padding explaining known concepts. Minor repetition — "不编造"/"接口未返回" stated three times and the production boundary restated several times — keeps it below a 5. | 4 / 5 |
Actionability | Copy-paste-ready commands with all flags (`douyin-hunter fetch --url … --video-only --download-media`, `viral-chaser --video … --output-dir …`), a complete output JSON schema, exit-code semantics, exact report table formats, and a verbatim DNA transcript template with the destination path. Specific examples cover the common cases per platform. | 5 / 5 |
Workflow Clarity | Steps 1–4 are clearly sequenced with explicit validation: documented exit codes 0/1/2, abort rules (hunter failure → do not run analyzer), required handling of warning flags (`KEY_FRAMES_PARTIAL`, `transcript.truncated`), and the checkpoint "只有转写与抽帧完成并核对结果后才写分析结论". | 5 / 5 |
Progressive Disclosure | Referenced bundle files are real and one level deep (viral-chaser wrapper → scripts/analyzer.ts; scripts/audio_extractor.ts, transcriber.ts exist), and the body is well-sectioned. The ~30-line inlined output JSON schema and the DNA sample template could live in references/ files, which is a minor organization gap. | 4 / 5 |
Total | 18 / 20 Passed |