Content
50%Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.
The content is well-organized into a clear 5-step workflow backed by concrete API commands and avoids explaining concepts Claude already knows. Its main weaknesses are malformed (non-executable) JSON bodies in the curl examples and a duplicated example section, both of which keep it at a solid-but-not-excellent 2 across dimensions.
Suggestions
Fix the malformed JSON bodies so each curl `-d` payload is a single valid object containing both the routing fields and `searchParams`/`query`, making the commands copy-paste executable.
Remove or merge the redundant "Example Usage" section that repeats the Step 1 search pattern, to tighten the token budget toward conciseness 3.
Add brief validation/error-handling notes (e.g., check for empty result sets or API error responses before proceeding to the next step) to add explicit checkpoints to the workflow.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The body is mostly lean curl commands with little explanatory padding and assumes Claude's competence, but the "Example Usage" section largely duplicates the Workflow's search patterns and the two-line credential extraction could be combined, so it could be tightened rather than earning the lean 3. | 2 / 3 |
Actionability | Concrete curl commands are provided throughout, but the JSON bodies are malformed as written (e.g., `-d '{...}'` closes the object before the dangling `"searchParams": {...}`), so the commands are not copy-paste executable and fall to "some concrete guidance but incomplete" rather than fully executable. | 2 / 3 |
Workflow Clarity | The five steps (search → research companies → company intel → find hiring managers → get contact info) are clearly sequenced, but there are no validation checkpoints or error/empty-result handling, so checkpoints are missing or implicit rather than explicit. | 2 / 3 |
Progressive Disclosure | The skill is a single ~130-line monolithic SKILL.md with no references/scripts/assets bundle, covering five API integrations all inline with section organization; it has structure but content that could be split out is inline, so it does not reach the well-signaled one-level-deep reference pattern of a 3. | 2 / 3 |
Total | 8 / 12 Passed |