Content
82%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.
A highly actionable, well-sequenced skill body with executable commands at every step, an explicit validation step, and a thoughtful human-approval gate plus untrusted-content handling before publishing. Main gaps are missing error-recovery loops after validation failure, repeated version pins, and the awkwardly placed TweetClaw subsection.
Suggestions
Add a feedback loop after step 4: what to do when 'socialclaw validate' fails (read the error, fix schedule.json, re-validate) before proceeding to apply.
Mention the socialclaw version pin once (e.g., in the Source section) and drop the duplicate in Setup to reduce time-sensitive detail.
Move the 'Optional X/Twitter evidence packet' subsection out of the build-schedule step — e.g., into its own section after Core Workflow or into Related Skills — so the numbered workflow stays uncluttered.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The body is efficient — commands, a compact provider table, and short security bullets with no explanations of concepts Claude already knows. It falls short of the score-5 anchor ('every token earns its place') due to repeated time-sensitive version pins ('socialclaw@0.1.12' appears twice, 'openclaw plugins install npm:@xquik/tweetclaw@1.6.31') that the guidelines say should penalize conciseness outside a deprecation section, plus the tangential TweetClaw evidence-packet paragraph. | 4 / 5 |
Actionability | Every step is copy-paste ready: 'socialclaw accounts list --json', 'socialclaw assets upload --file ./image.png --json', a complete schedule.json example with expected outputs ('→ { "run_id": "..." }'). This matches the score-5 anchor — fully executable commands covering the common cases — and exceeds score 4, which allows minor gaps. | 5 / 5 |
Workflow Clarity | The six-step Core Workflow has a clear sequence with an explicit validation checkpoint (step 4 'Validate before publishing') and a strong approval gate before apply ('show the user every post... and get explicit approval'). It stops short of the score-5 anchor, which requires feedback loops for error recovery — there is no 'if validation fails, fix and re-validate' instruction. | 4 / 5 |
Progressive Disclosure | No bundle files exist (references/, scripts/, assets/ are absent), so all content lives in a single well-sectioned SKILL.md at overview level, with nothing that clearly demands a separate file. It matches the score-4 anchor ('good structure; most content appropriately placed; minor organization gaps') rather than 5 because the 'Optional X/Twitter evidence packet' subsection sits oddly inside the build-schedule step and slightly breaks the otherwise clean navigation. | 4 / 5 |
Total | 17 / 20 Passed |