Content
57%Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.
This skill provides a solid mathematical framework for social graph ranking with clear formulas and a useful output template. Its main weaknesses are the lack of executable code or concrete tool commands for actually performing the computation, and the absence of validation checkpoints in the workflow. The conceptual model is strong but the skill reads more like a specification than an actionable instruction set.
Suggestions
Add executable code (Python or pseudocode with clear library references) for computing B(m), B_ext(m), and R(m) from actual graph data structures, making the math directly implementable.
Add validation checkpoints to the workflow, e.g., verify graph data completeness after step 2, sanity-check score distributions after step 5, and handle edge cases like disconnected targets.
Trim the 'Scoring Signals' section significantly—most of these weighting factors (role alignment, geographic relevance, etc.) are obvious to Claude and don't need enumeration.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The skill is reasonably efficient but includes some unnecessary sections like 'When To Use This Standalone' with example phrases that could be trimmed, and the 'Scoring Signals' section largely restates obvious weighting factors Claude would already know. The math notation is appropriately dense, but surrounding prose could be tighter. | 2 / 3 |
Actionability | The math formulas are concrete and well-defined, and the output shape provides a clear template. However, there's no executable code for actually computing the scores, no concrete commands for pulling graph data, and the workflow steps like 'Pull the user's graph from X, LinkedIn, or both' are vague about how to actually do this. The skill describes a model more than it instructs on execution. | 2 / 3 |
Workflow Clarity | The 6-step workflow is clearly sequenced and logical, but lacks any validation checkpoints. There's no verification step after pulling graph data, no check on whether the target set is well-formed, and no feedback loop for when graph data is incomplete or bridge scores seem anomalous. For a multi-step computational workflow, this is a notable gap. | 2 / 3 |
Progressive Disclosure | The skill is well-structured with clear sections progressing from context to model to workflow to output. Related skills are clearly signaled at the end with one-level-deep references to other skills. The content is appropriately self-contained for the ranking engine while pointing elsewhere for broader workflows. | 3 / 3 |
Total | 9 / 12 Passed |