Content
82%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 tight, highly actionable skill body: executable examples dominate, guidance is non-obvious where it needs to be (the .shape gotcha, env vars, in-process fetch), and structure is easy to navigate. The only weaknesses are a missing verification step in the scaffold workflow and modest reference material that could be split out.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The body is efficient — file locations, key rules, and executable snippets with no padding about concepts Claude already knows. Anchor 5 ('every token earns its place') isn't quite met: the Anthropic-compatibility table and the Key rules list partly restate what the code comments already show ('NOTE: .shape, not the ZodObject'). | 4 / 5 |
Actionability | Fully executable, copy-paste-ready code: the minimal plugin shape, a complete two-tool example with system-prompt hook, exact opencode.json registration snippets, and a concrete scaffold workflow. These cover the common cases, matching anchor 5; it is well above anchor 4's 'minor gaps'. | 5 / 5 |
Workflow Clarity | The 'Scaffold workflow' gives a clear numbered sequence (create the plugin file, add to opencode.json, restart session). Anchor 5 is not met because there is no verification checkpoint (e.g., confirming the tool loads or the plugin compiles), but this is not a destructive/batch operation, so no cap applies. | 4 / 5 |
Progressive Disclosure | Well-sectioned with clear headers (When to use, Plugin API, hooks, registration, compat, example, workflow) and no bundle files exist, so everything lives appropriately in SKILL.md with easy navigation — matching anchor 4. It falls short of anchor 5 only because reference-style material (the hooks API listing and the Anthropic compatibility section) is inlined where a references/ split could keep SKILL.md leaner. | 4 / 5 |
Total | 17 / 20 Passed |