Content
58%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 an extremely lean stub whose entire content is a redirect to an external skill file outside the bundle. While token-efficient and unambiguous as an instruction, it provides no actual task guidance, no workflow, and its single reference cannot be verified from the skill's own files, leaving a reader dependent on an unresolvable path.
Suggestions
Inline a minimal quick-start (how to list/find aven tasks and update status) instead of delegating 100% of the guidance to an external path that is not part of the skill bundle.
If delegation is intentional, reference a file that ships inside the skill bundle (e.g. references/SKILL-core.md) so the pointer is resolvable and verifiable, and briefly state what the reader will find there.
Add at least a one-line summary of the core workflow (find task → update status → leave handoff note) so the skill body stands on its own if the external file is unavailable.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The body is two lines — an HTML generation comment and "Load and follow the shared skill at `~/.agents/skills/aven/SKILL.md`" — with zero padding or explanation of concepts Claude already knows. Every token present is doing work; this matches the "lean and efficient; every token earns its place" anchor exactly. | 5 / 5 |
Actionability | "Load and follow the shared skill at `~/.agents/skills/aven/SKILL.md`" is a concrete, specific directive with a real path, but the body itself contains none of the actual task guidance — no commands, examples, or steps for finding tasks, updating status, or leaving handoff context. Not 4 because the executable guidance is only the redirect itself, leaving major gaps; not 2 because the one instruction given is specific and unambiguous rather than a high-level hint. | 3 / 5 |
Workflow Clarity | The only step present is a single delegation instruction; the workflows for the capabilities named in the description (find tasks, update status, write handoff context) are entirely absent, with no sequencing and no validation checkpoints. The simple-skill exception does not apply because the description covers multiple capabilities, not one unambiguous single task. Not 3 because there is no sequence of steps listed at all beyond the redirect — only a degenerate one-step body with large gaps. | 2 / 5 |
Progressive Disclosure | The single reference is clearly signaled and one level deep, but it points outside the skill bundle to an absolute home-directory path (`~/.agents/skills/aven/SKILL.md`) rather than to files in references/, and no bundle files exist to verify. The body provides no overview content of its own — it is a bare redirect, which is "some structure" but not well-organized progressive disclosure. Not 4 because content placement cannot be assessed and the reference is unresolvable from the bundle; not 2 because the reference is not buried and there is no inlined content that belongs elsewhere. | 3 / 5 |
Total | 13 / 20 Passed |