Content
57%Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.
This is a well-structured coordination skill with a clear mental model (GitHub = public, Linear = internal) and a useful classification framework. Its main weakness is lack of actionability—it describes what to do conceptually but never specifies which tools to call or how to invoke them, making it more of a decision guide than an executable skill. Adding specific tool invocations and a validation step would significantly improve it.
Suggestions
Add specific tool invocations for each workflow step (e.g., exact MCP tool names or CLI commands for reading GitHub PRs, creating Linear issues, posting comments back to GitHub).
Include a validation/verification step after Linear creation and GitHub comment posting to confirm the action succeeded and the two systems are consistent.
Add a concrete worked example showing a real PR triage from raw input through classification to the final output format, demonstrating the decision logic in action.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | Generally efficient but includes some unnecessary elaboration. The 'When to Use' section partially overlaps with the description, and some bullet lists could be tightened. The 'Operating Model' section explains concepts (GitHub is public, Linear is internal) that could be stated more tersely, though the decision criteria for Linear creation add genuine value. | 2 / 3 |
Actionability | Provides a clear classification framework and output format template, which are concrete and useful. However, there are no executable code examples, no specific tool invocations (e.g., actual GitHub CLI commands, Linear API calls, or MCP tool names), and the workflow steps are more descriptive than prescriptive. 'Gather GitHub issue or PR state' doesn't tell Claude how. | 2 / 3 |
Workflow Clarity | The four-step workflow is clearly sequenced and the classification table is well-structured. However, there are no validation checkpoints or feedback loops—no step to verify that the Linear item was correctly created, no check that the GitHub comment was posted, and no error recovery guidance if tool calls fail. | 2 / 3 |
Progressive Disclosure | For a skill of this size (~80 lines) covering a single coordination workflow, the content is well-organized with clear sections (When to Use, Operating Model, Core Workflow, Review Rules, Output Format, Good Use Cases). No external references are needed and the structure supports easy scanning. | 3 / 3 |
Total | 9 / 12 Passed |