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.
The body is a maximally lean delegation stub: perfectly concise and clearly signaled, but it contains none of its own task guidance — the command, timing, and verification logic all live in an external file not shipped with the bundle. It works as a redirect but is not self-contained.
Suggestions
Inline the core directive in the body (e.g., run `cmux workspace status set done` when the session goal is reached and verified) so the skill remains actionable even if the shared skill at ~/.agents/skills/cmux-workspace-status/SKILL.md is unavailable.
Add a fallback or validation step for the delegation, such as checking that the shared skill file exists before relying on it, or stating what to do when it is missing.
Fix the dangling references: point to bundle-relative paths where possible and remove or correct the mention of scripts/agent-harnesses.py, since no scripts/ directory exists in this bundle.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The body is two lines — a generated-by marker and "Load and follow the shared skill at `~/.agents/skills/cmux-workspace-status/SKILL.md`" — with zero padding and nothing Claude doesn't need. Every token earns its place; this matches the lean anchor 5 exactly. | 5 / 5 |
Actionability | The single instruction is concrete (an exact path to load and follow), but the executable guidance for the skill's actual task — the `cmux workspace status set done` command, when to run it, how to verify the goal — is entirely absent from the body and delegated to an external file outside the bundle. This fits anchor 3 (concrete but incomplete, missing key details) better than anchor 4, which expects actual code or commands in the content. | 3 / 5 |
Workflow Clarity | As a simple single-task skill the one action ("Load and follow the shared skill...") is unambiguous, which supports a high score under the simple-skill exception. However the body provides no checkpoints or fallback for the delegation itself — nothing to do if the shared skill file is absent — and no steps for the underlying status-setting operation, leaving minor validation gaps characteristic of anchor 4 rather than a clean 5. | 4 / 5 |
Progressive Disclosure | The reference is clearly signaled and one level deep (no nested chains), and nothing that belongs in a separate file is inlined. It stops short of anchor 5 because the pointer targets an absolute path outside the bundle (`~/.agents/skills/...`) rather than a bundle-relative file, and the header comment cites `scripts/agent-harnesses.py` which does not exist in the bundle (no references/, scripts/, or assets/ directories are present). | 4 / 5 |
Total | 16 / 20 Passed |