Content
85%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 well-structured, highly actionable skill body with strong verification checkpoints, error-recovery fallbacks, and clean one-level-deep platform references. The weaknesses are local: a subtle printf bug in the capture example, one externally deferred command in section C, and a few trimmable rationale sentences.
Suggestions
Fix the flow A example: `printf '%s' "## <title>\n\n1. ..."` will not expand \n in the argument (printf only processes escapes in the format string). Use `printf '%b'`, put the newlines in the format string, or pipe a heredoc to `--body-file -`.
In section C, inline the exact transcript-drill command (or a one-line example of it) instead of deferring to the memory-recall skill's platform reference, so the flow is self-contained.
Trim rationale sentences that restate the instruction (e.g. "so the result can be more accurate", "a wrong command is worse than a vague one") where the directive alone already changes behavior.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The body is dense and assumes competence — no explanations of concepts Claude already knows — but a few rationale sentences could be trimmed, e.g. "doing it here on demand lets you inspect the original transcripts more deliberately, so the result can be more accurate" and "a wrong command is worse than a vague one". This matches anchor 4 ('efficient; minor instances of over-explanation that could be trimmed') rather than anchor 5's 'every token earns its place'. | 4 / 5 |
Actionability | Nearly every step ships a copy-paste command (`memsearch skills add --name ... --body-file -`, `skills status`, `skills install <name> --path ...`, concrete `config get/set` lines with example values). Two minor gaps keep it below anchor 5: section C defers the transcript-drill command to "the memory-recall skill's platform reference" instead of stating it, and the flagship `printf '%s' "...\n..."` example is subtly broken — printf's %s does not expand \n in the argument, so the body-file would contain literal backslash-n sequences. | 4 / 5 |
Workflow Clarity | The stage model (0→1→2), intent routing, and per-flow step order are explicit, and validation is built into each flow: "Before recommending or installing, skim the candidate's body: if a step looks uncertain... re-check it against the source", "Treat installation as an interactive checkpoint... confirm the install destination with the user", "Be exact — do not guess... verify it", and an explicit error-recovery fallback "Only if that command fails (unknown format) fall back to reading the raw file directly". | 5 / 5 |
Progressive Disclosure | The SKILL.md body is a clear overview (stages, routing, flows, guardrails) with platform-specific detail split into five real, one-level-deep reference files (references/claude-code.md, codex.md, openclaw.md, opencode.md, dsh.md — all present, none nesting further references), each clearly signaled by a platform bullet list and 'See your platform reference file' pointers. | 5 / 5 |
Total | 18 / 20 Passed |