Content
75%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 lean, well-structured methodology document that gives concrete structure, packaging, and workflow guidance with validation checkpoints and a sensible fallback when the external methodology file is absent. Its gaps are the absence of any example CLI command, an unspecified TEST.md format, no explicit fix-and-retry loop, and a primary reference that is external to the bundle rather than shipped in references/.
Suggestions
Add one short example of a generated subcommand (Click decorator, --json flag, REPL behavior) so implementers have copy-paste-ready guidance for the core deliverable.
Make the test workflow a feedback loop: after step 5, add 'If tests fail, fix the harness and re-run before proceeding' to reach explicit validation/recovery.
Ship the condensed HARNESS.md methodology as a references/ file inside the bundle (linking it from SKILL.md) so the primary source of truth survives standalone distribution instead of relying on ../cli-anything-plugin/.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The body is tight and rule-driven with no explanations of concepts Claude already knows, but "Use this skill when the user wants OpenClaw to act like the `CLI-Anything` builder." duplicates the frontmatter description and could be trimmed. | 4 / 5 |
Actionability | Concrete guidance throughout — an exact directory layout, specific filenames (test_core.py, utils/<software>_backend.py), exact packaging directives (find_namespace_packages(include=["cli_anything.*"]), console_scripts -> cli-anything-<software>), and `pip install -e .` — but no example CLI command or signature and no specification of TEST.md's content. | 4 / 5 |
Workflow Clarity | The 7-step numbered Workflow is clearly sequenced with validation checkpoints ("Write TEST.md, then tests, then run them"; "Verify local installation with `pip install -e .`") plus a Validate-mode checklist, but there is no explicit error-recovery loop (e.g., 'if tests fail, fix and re-run'). | 4 / 5 |
Progressive Disclosure | Well-sectioned content with a single, clearly signaled one-level-deep reference and a graceful fallback ("If it is not available, follow the condensed rules below"), but the sole reference (`../cli-anything-plugin/HARNESS.md`) lives outside the skill bundle, so it dangles when the skill ships standalone. | 4 / 5 |
Total | 16 / 20 Passed |