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 content is a strong operational skill: fully executable commands, real log strings to seed filters, and a status-first workflow with explicit profile/freshness checkpoints. The only weaknesses are minor duplication between the applicability table and later sections, and the entire target/span cheat-sheet being inlined in SKILL.md rather than split into a reference file. A stray unclosed ```text fence at the end of the file should also be cleaned up.
Suggestions
Split the subsystem target/span cheat-sheet (and optionally the jq cookbook) into a references/ file (e.g. references/cheatsheet.md) and point to it from SKILL.md, keeping the main file to the layout, commands, and workflow.
Bundle the uc-logs.sh helper script inside the skill (e.g. scripts/uc-logs.sh) or note that it lives in the repo, since the skill names it as the sole required entry point but it is absent from the skill bundle.
Remove the duplication between the applicability table and the 'Relationship to export-logs' / legacy-layout sections, and delete the stray unclosed ```text fence at the end of the file.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The body is almost entirely project-specific knowledge Claude cannot know (per-platform log dirs, UC_PROFILE resolution, real target/span strings) and assumes competence, never explaining JSONL, jq, or UTC. It falls short of the 5 anchor only through minor duplication — the legacy-layout warning and the export-logs guidance each appear in both the applicability table and a later section, and a little workflow prose could be tightened. | 4 / 5 |
Actionability | Everything is copy-paste executable: the `uc-logs.sh <command>` table with flags and defaults, ready-to-use jq filters seeded with real field names and example ids (`.request_id == "b6c8588aa8a8699c"`), a compact projection pipeline, and `uniclip debug export-logs [--since-hours N]`. Specific examples cover the common diagnostic cases. | 5 / 5 |
Workflow Clarity | The workflow has explicit validation with error recovery: run `status` first, interpret per-role freshness (live/recent/stale/cold), and "stop and confirm the profile… Don't silently analyze stale logs" if the relevant role is cold, followed by a clear 5-step sequence (ground → narrow time window → broad-then-narrow → correlate by id → quote sparingly). This is a read-only diagnostic skill, so the destructive/batch validation cap does not apply. | 5 / 5 |
Progressive Disclosure | A single well-organized file with clear section headers and one clearly-signaled, one-level-deep external path (`uc-logs.sh`). It stops short of the 5 anchor because the bundle ships no reference/script files at all — the skill depends on an external script not bundled here — and the ~40-line target/span cheat-sheet and jq cookbook are natural candidates for a separate reference file to keep SKILL.md lean. | 4 / 5 |
Total | 18 / 20 Passed |