Content
72%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 highly actionable with complete, executable CLI commands and a well-sequenced connection flow, but it is a monolithic single file with inlined reference tables and lacks validation/verification steps for the destructive and batch action operations it supports.
Suggestions
Add a validation/verification checkpoint to the action-running workflow, e.g. re-fetching or inspecting the `output` field to confirm destructive operations (delete-subscription, send-message) succeeded before proceeding.
Move the 15-row Popular actions table and the proxy-flags table into a separate `references/actions.md` file, keeping SKILL.md a concise overview with a clearly signaled one-level-deep reference.
Trim the filler 'Use action names and parameters as needed.' line and tighten the opening overview sentence to remove marketing-style padding.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The body is mostly efficient with concrete commands and tables, but contains minor trimmable padding such as the marketing-style overview sentence and the filler line 'Use action names and parameters as needed.' | 4 / 5 |
Actionability | Provides fully executable, copy-paste-ready bash commands across the common cases (install, login, connection ensure, action list/run with JSON input, proxy) plus concrete action keys and flag tables. | 5 / 5 |
Workflow Clarity | The connection sub-workflow has explicit state checkpoints and a poll-retry loop, but the core action-running workflow (covering destructive/batch operations like delete-subscription and send-message) lacks any validation or verification step, which caps the score at 3 per the destructive-operations guideline. | 3 / 5 |
Progressive Disclosure | The skill is a single monolithic ~160-line file with reference-style content (a 15-row popular-actions table and a proxy-flags table) inlined that could appropriately live in a separate references file, with no one-level-deep navigation to deeper material. | 3 / 5 |
Total | 15 / 20 Passed |