Content
96%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 an exemplary instruction-only skill: lean, fully concrete, and a well-gated nine-step workflow with explicit validation and error recovery around the destructive archive operation. The only minor weakness is navigation polish around template placeholders.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The ~78-line body contains no padding and no explanations of concepts Claude already knows; every sentence is operational (file paths, gating conditions, CLI consequences). It assumes competence and each token earns its place, matching the lean anchor. | 5 / 5 |
Actionability | Guidance is fully executable: specific paths (.apm/spec.md, .apm/session.summary.md, .apm/archives/index.md), copy-paste commands ("apm archive --force --name <name>", "apm init"), and complete markdown templates for both output files. Per the code_vs_instruction note, this instruction-only skill has concrete, specific guidance throughout. | 5 / 5 |
Workflow Clarity | Nine clearly numbered steps with explicit validation checkpoints: step 2 gates on confirmed Worker state ("ask the User to pause them"), and step 7 gates the destructive archive on CLI confirmation with a failure-recovery feedback loop ("If it reports a failure, show the obstacle... and follow its recovery guidance"). This matches the anchor with explicit validation and error-recovery loops. | 5 / 5 |
Progressive Disclosure | No bundle files exist, and the body is cleanly organized (numbered workflow, subagents subsection, separate formats section) with one clearly signaled one-level reference ("Read {SKILL_PATH:apm-communication}"). The unresolvable template placeholders ({VERSION}, {SKILL_PATH:...}) are not clickable links, a minor navigation gap that keeps it below anchor 5. | 4 / 5 |
Total | 19 / 20 Passed |