Content
63%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 read-only analysis workflow with a concrete prerequisite gate and precise output specification. Its main weaknesses are redundancy — the hook protocol duplicated verbatim and a repeated $ARGUMENTS section — and the absence of any bundle files to offload that detail.
Suggestions
Deduplicate the extension-hook protocol: state the shared rules once and have the pre/post sections reference them (e.g., 'apply the hook procedure from Pre-Execution Checks with hooks.after_analyze'), or move the hook spec into a references/hooks.md file.
Remove the redundant trailing '## Context / $ARGUMENTS' section (the 'User Input' section at the top already covers it) and drop the stray shell-quoting tip ('I'm Groot') that belongs to the prerequisite-script context, not this skill.
Make the coverage mapping step concrete: specify the exact matching procedure (e.g., 'match FR-/SC- IDs appearing in task descriptions or referenced file paths; list unmatched IDs in the Coverage Summary Table') so the analysis is reproducible.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The body is mostly efficient and procedural, but the extension-hook protocol is duplicated nearly verbatim across the pre-execution and post-analysis sections (~80 combined lines), '$ARGUMENTS' appears in two sections ('User Input' at top and 'Context' at bottom), and the stray shell-quoting tip ("I'm Groot") is padding. This matches anchor 3 ('Mostly efficient but includes some unnecessary explanation or could be tightened') rather than anchor 4. | 3 / 5 |
Actionability | Concrete, executable guidance throughout: an exact script invocation ('.specify/scripts/bash/check-prerequisites.sh --json --require-spec --require-tasks --include-tasks'), derived absolute paths, a fully specified report format with table columns and an example row, and explicit severity heuristics. It stops short of anchor 5 because steps like the task-coverage mapping are described abstractly ('inference by keyword / explicit reference patterns') without concrete executable detail. | 4 / 5 |
Workflow Clarity | A clear 9-step sequence with an explicit validation checkpoint (prerequisite check that aborts with instructive error messaging) and graceful zero-issues handling; the operation is strictly read-only, so the destructive/batch validation cap does not apply. It is not anchor 5 because there is no step verifying the report's internal self-consistency (e.g., that coverage metrics match the tables) — a minor validation gap. | 4 / 5 |
Progressive Disclosure | Sections are well-organized with clear headers, but at ~250 lines the duplicated hook machinery (emitted twice, pre and post) is content that clearly belongs in a shared reference file, matching anchor 3 ('Some structure... content that should be separate is inline'). It does not reach anchor 4 given this inlined, repeated bulk with no external references at all. | 3 / 5 |
Total | 14 / 20 Passed |