roborev-fix
Use when the user asks to fix open reviews, invokes /roborev-fix, or provides job IDs; do not use when the user only pastes review findings with no request to discover or close reviews
68
83%
Does it follow best practices?
Impact
—
No eval scenarios have been run
Advisory
Suggest reviewing before use
Quality
Discovery
89%Based on the skill's description, can an agent find and select it at the right time? Clear, specific descriptions lead to better discovery.
This description excels at trigger clarity and distinctiveness, with explicit positive and negative use-case boundaries and a unique slash command trigger. Its main weakness is that the 'what does this do' component is underdeveloped—it says 'fix open reviews' but doesn't elaborate on the specific actions or capabilities involved (e.g., querying review systems, applying fixes, closing tickets).
Suggestions
Add explicit capability statements describing what the skill does, e.g., 'Queries open code reviews by job ID, applies automated fixes, and closes resolved review items.'
| Dimension | Reasoning | Score |
|---|---|---|
Specificity | The description mentions 'fix open reviews' and 'discover or close reviews' which hint at concrete actions, but it doesn't list specific capabilities comprehensively—what does 'fixing' entail? The actions remain somewhat vague beyond the domain context. | 2 / 3 |
Completeness | The description explicitly answers both 'when to use' (user asks to fix open reviews, invokes /roborev-fix, provides job IDs) and 'when NOT to use' (only pastes review findings with no request to discover or close). The 'what' is somewhat implicit but the 'when' guidance is exceptionally clear with both positive and negative triggers. | 3 / 3 |
Trigger Term Quality | Includes strong natural trigger terms: 'fix open reviews', '/roborev-fix', 'job IDs', 'review findings', 'discover or close reviews'. These are terms a user would naturally use, and the slash command is a clear explicit trigger. | 3 / 3 |
Distinctiveness Conflict Risk | The description is highly distinctive with the specific '/roborev-fix' command, 'job IDs', and the explicit negative boundary condition about pasted review findings. This creates a clear niche unlikely to conflict with other skills. | 3 / 3 |
Total | 11 / 12 Passed |
Implementation
77%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-crafted, highly actionable skill with excellent workflow clarity including validation checkpoints, error recovery, and an explicit audit step. Its main weakness is verbosity — the skill is thorough but could be more concise, particularly around edge case handling and conditional logic that could be compressed. The monolithic structure is acceptable given no bundle exists, but the length suggests some content could be extracted into reference files.
Suggestions
Compress the 'When NOT to invoke this skill' section and the conditional logic in step 1 (pasted findings vs. fetched vs. discovered) into a more concise decision table or flowchart format.
Consider extracting the JSON output structure documentation and panel review handling details into a separate reference file to reduce the main skill's token footprint.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The skill is thorough but somewhat verbose. Several sections explain edge cases and conditional logic at length (e.g., panel reviews, synthesis parents, pasted findings handling). Some of this detail earns its place given the complexity, but sections like the 'When NOT to invoke this skill' block and repeated explanations of skip conditions could be tightened. The skill assumes Claude's competence in some areas but over-explains in others. | 2 / 3 |
Actionability | The skill provides fully executable bash commands throughout (roborev show, roborev fix --list, roborev comment, roborev close, git show, go test). Each step has concrete commands with specific flags and arguments. The JSON output structure is documented, and decision logic (skip passing reviews, handle errors) is explicit and actionable. | 3 / 3 |
Workflow Clarity | The 7-step workflow is clearly sequenced with explicit validation checkpoints: test verification in step 4, confirming comment success before closing in step 5, and a dedicated audit step (step 7) to verify closures. Error recovery is addressed (report errors and continue, fix regressions before proceeding). The closure ordering mandate and feedback loops (fix -> test -> close -> audit) are exemplary. | 3 / 3 |
Progressive Disclosure | The skill is a single monolithic file with no bundle files to reference. While it does reference CLAUDE.md and /roborev-respond appropriately, the content is quite long (~200+ lines) and could benefit from splitting detailed JSON schema documentation or panel review handling into separate reference files. The examples section is well-placed but the main instructions section is dense. | 2 / 3 |
Total | 10 / 12 Passed |
Validation
100%Checks the skill against the spec for correct structure and formatting. All validation checks must pass before discovery and implementation can be scored.
Validation — 11 / 11 Passed
Validation for skill structure
No warnings or errors.
- Repository
- roborev-dev/roborev
- Commit
a407fdb
- Reviewed
Table of Contents
Is this your skill?
If you maintain this skill, you can claim it as your own. Once claimed, you can manage eval scenarios, bundle related skills, attach documentation or rules, and ensure cross-agent compatibility.