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 body is a tightly written, command-driven workflow with strong validation checkpoints, explicit authorization gates for risky actions, and clean hand-off boundaries to sibling skills. Its main quality lever is trimming a few redundant constraint restatements. Progressive disclosure is solid for a single-file skill but relies on external references that are not part of this bundle.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The body is lean, imperative, and assumes competence — e.g., 'Refresh the trusted origin/main reference, then compare the entire checkout/worktree diff' with no concept explanations or padding. It sits at 'Efficient; minor instances that could be trimmed' rather than 5: lines like 'Do not duplicate its dependency, build, hook, CLI-exposure, or readiness logic in agent commands' and the repeated setup/repair scoping in 'Reserve setup and --repair for repository-local dependency, build, or hook repair' restate earlier constraints. | 4 / 5 |
Actionability | Every route maps to a copy-paste command: './scripts/dev-setup.sh --doctor', '--doctor --json', '--repair', '--expose-cli', '--with-runtime', 'gh auth login -h github.com', 'npm run agent', plus 'Pass user-supplied Pi arguments after --'. Commands cover all the common cases the routing section names, matching 'Fully executable; copy-paste ready code or commands; specific examples cover the common cases'. | 5 / 5 |
Workflow Clarity | Clear sequence (establish trust → route by intent → handle user-controlled changes → runtime decision → pinned agent → handoff → report) with explicit validation checkpoints and feedback loops: 'run the doctor first', 'If readiness fails, report the remediation and obtain authorization... rather than mutating the checkout automatically', and 'rerun npm run dev:doctor... instead of rerunning setup' after remediation. This matches the anchor for explicit validation steps with error-recovery loops; risky execution surfaces are gated behind diff review and approval. | 5 / 5 |
Progressive Disclosure | Good structure with one clearly signaled one-level-deep reference ('Follow ../_shared/git-github-hard-stop.md for authentication, authorization, SSH, remote-access, or push failures') and hand-off pointers naming sibling workflows instead of restating them. It matches 'Good structure; most content appropriately placed; references mostly clear' rather than 5 because the skill is ~90 lines of fully inline content with no bundle of its own (no references/, scripts/, or assets/ files exist here), so some scenario detail (e.g., the git-github hard-stop rules) necessarily lives outside with no local navigation surface. | 4 / 5 |
Total | 18 / 20 Passed |