Content
46%Scale 1-5Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.
This skill provides a comprehensive but overly verbose guide to GitHub issue management via the gh CLI. Its main strength is the concrete CLI commands and structured body templates for different issue types. Its main weaknesses are excessive verbosity (especially the repeated AskUserQuestion specifications and title guidelines), lack of validation checkpoints in the workflow, and a monolithic structure that would benefit from splitting content across files.
Suggestions
Dramatically reduce the AskUserQuestion specifications — Claude doesn't need every option spelled out in full detail. Summarize the interaction flow and focus on the gh CLI commands and body templates.
Split the three action paths (create, list, view) and the issue body templates into separate referenced files to improve progressive disclosure and reduce the main file's length.
Add inline validation after gh commands — e.g., check the exit code, confirm the issue URL was returned, and handle the case where the command fails before reporting success.
Remove redundant title guidelines (good/bad examples are repeated) and description guidelines that explain obvious concepts like 'include context' and 'be thorough.'
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The skill is significantly verbose, with extensive step-by-step UI interaction scripts that over-specify AskUserQuestion parameters. The title guidelines section repeats good/bad examples unnecessarily, and much of the content reads like a tutorial rather than concise instructions for Claude. The description guidelines and multiple question steps for bug reports add substantial bulk. | 2 / 5 |
Actionability | The skill provides concrete gh CLI commands for each action (create, list, view) and includes executable bash examples including the HEREDOC pattern. However, the actionability is somewhat diluted by the heavy focus on UI interaction scaffolding (AskUserQuestion specifications) rather than the actual execution logic. The commands themselves are clear and copy-paste ready. | 4 / 5 |
Workflow Clarity | The multi-step workflows are clearly sequenced with numbered steps and branching paths for each action type. However, there are no validation checkpoints — the skill doesn't verify that the gh command succeeded before reporting success, and the error handling section is an afterthought rather than integrated into the workflow. For a skill that creates issues (a non-destructive but external operation), some inline verification would strengthen it. | 3 / 5 |
Progressive Disclosure | All content is inlined in a single monolithic file with no references to supporting files. The three action paths (create, list, view) and the bug report template could easily be split into separate reference files. The issue body templates alone would benefit from being in a separate file. With no bundle files provided, the skill dumps everything into one long document. | 2 / 5 |
Total | 11 / 20 Passed |