Content
86%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.
An exceptionally lean, imperative body whose central command is executable as written, with clear sequencing and safety gates for risky git operations. The main weakness is the unnamed "required task-start script" in step 1, which is not directly executable.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The 6-step body is lean with zero padding — no concept explanations, no filler, every line directs behavior. It assumes Claude's competence (e.g., "Inspect branch, upstream, worktrees, status" without re-explaining git), matching the level 5 anchor exactly. | 5 / 5 |
Actionability | Step 2 is a fully copy-paste-ready command ("npm run workflow:lifecycle -- --phase <status|start|handoff|landed|cleanup> --write-evidence") and steps 4–6 are concrete directives. It falls short of 5 because step 1's "required task-start script" is never named or given a path, and step 3 lists what to inspect without the commands to do so — minor but real executable gaps. | 4 / 5 |
Workflow Clarity | A clearly sequenced 6-step process with most checkpoints present: inspection of branch/upstream/worktrees/markers (step 3), approval gates on risky operations (step 5), and an explicit report of the proved transition (step 6). It is not a 5 because there is no error-recovery feedback loop (what to do when inspection reveals a problem), only implicit checkpoints. | 4 / 5 |
Progressive Disclosure | The skill is under 50 lines with no bundle files (references/, scripts/, assets/ are absent) and no content that warrants external files, so per the simple-skill guideline a well-organized single body scores 5. The single header with a flat numbered list is clean and fully navigable. | 5 / 5 |
Total | 18 / 20 Passed |