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 a tight, actionable, well-sequenced workflow with executable commands and explicit validation/feedback loops. The only sub-5 dimension is progressive disclosure, where splitting the script-format detail or fraimz-handoff into a reference file would push it to 5.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | Lean body that assumes Claude's competence (no explanation of worktrees, PRs, or scaffolding) with every section earning its place; only a couple of framing sentences are mildly narrative. | 5 / 5 |
Actionability | Copy-paste-ready bash commands (git worktree add, pnpm fraimz scaffold, gh pr create, pnpm fraimz --flow --pr), concrete file paths, and an explicit script-format example cover the common cases. | 5 / 5 |
Workflow Clarity | A clear four-phase numbered sequence with explicit validation checkpoints and a feedback loop: "no code until the script is approved", the runner failing on narration drift, and "repair and re-run until every frame passes". | 5 / 5 |
Progressive Disclosure | Well-organized into phases with one-level-deep references to source-of-truth files and the fraimz skill; at ~70 lines it slightly exceeds the under-50-line simple-skill threshold, so it stays just below 5 rather than at it. | 4 / 5 |
Total | 19 / 20 Passed |