Content
77%Reviews 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 clear, validated workflows built around real CLI/MCP gotchas. Its main weakness is progressive disclosure: a single long file where reference-style tables could be split into bundle files, plus minor length-related conciseness concerns.
Suggestions
Move stable lookup tables (labels, story-point scale, custom-field mappings, resolution values) into a references/ file and link to it from SKILL.md so the main file stays a lean overview.
De-duplicate the closing guidance — required-close fields and the 'close from Internal/External Review' note are restated across the Ticket Lifecycle and Closing Issues sections.
Consolidate the trailing scattered flag notes (--plain, --raw, --csv, etc.) into the Non-Interactive Mode or CLI Gotchas section to reduce loose fragments.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The body is dense and almost entirely non-obvious operational knowledge (CLI gotchas, project-specific field keys, lifecycle), but at roughly 480 lines with some repetition around closing procedures it is longer than necessary; it is not level 3 because not every token earns its place, and not level 1 because it avoids explaining concepts Claude already knows. | 2 / 3 |
Actionability | Provides fully executable CLI bash and MCP python examples with specific flags, field keys, and copy-paste-ready snippets across create/edit/transition/comment/link operations; it is not level 2 because the guidance is concrete and complete rather than pseudocode. | 3 / 3 |
Workflow Clarity | Multi-step processes are clearly sequenced with validation checkpoints and feedback loops — tool selection with quick-fail fallback, closing issues with 'read the error for valid transition names' and 'use jira_search_fields' recovery, and a numbered scope-creep workflow with a confirmation step. | 3 / 3 |
Progressive Disclosure | Content is well-organized into clear sections, but everything lives inline in one ~480-line file with no bundle references; large lookup tables (labels, story points, field mappings) that could be split into reference files are inline, so it is not level 3, though it avoids being a disorganized wall of text. | 2 / 3 |
Total | 10 / 12 Passed |