CtrlK
BlogDocsLog inGet started
Tessl Logo

roborev-refine

Use only when the user explicitly invokes /roborev-refine

58

Quality

73%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide
SecuritybySnyk

Passed

No findings from the security scan

Fix and improve this skill with Tessl

tessl review fix ./internal/skills/grok/roborev-refine/SKILL.md
SKILL.md
Quality
Evals
Security

Quality

Content

88%Weight 40%Scale 1-5

Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.

The body is an exemplary operational skill: every step is executable, validation and feedback loops are explicit, and edge cases (panel parent jobs, hook reviews, sandbox escalation, default-branch refusal) are handled concretely. The only weaknesses are minor redundancy — the duplicated --since snippet and restated examples — that keep conciseness and progressive disclosure just below top marks.

DimensionReasoningScore

Conciseness

The body is dense and operational with no concept explanations Claude already knows, but the since-scoped bash snippet is duplicated verbatim in section 2 and section 3d, and the Examples section restates loop steps already specified in the Instructions. This matches anchor 4: efficient with minor over-explanation that could be trimmed.

4 / 5

Actionability

Fully executable throughout: copy-paste-ready bash with heredoc-safe variable passing (roborev comment via ROBOREV_COMMENT), exact commands (roborev review --since/--branch --wait, roborev wait, roborev close), exit-code handling for --wait, and two worked examples with clearly marked placeholders. This matches anchor 5's copy-paste-ready commands covering common cases.

5 / 5

Workflow Clarity

Clear numbered sequence with validation checkpoints at every risky step: branch-match verification, git rev-parse/merge-base --is-ancestor resolution of --since, tests before proceeding, comment-success-before-close ordering, and an explicit full-scope re-review gate before declaring success. The fix-to-re-review feedback loop with iteration limit and remaining-findings reporting matches anchor 5.

5 / 5

Progressive Disclosure

The skill is a single self-contained file with no references/, scripts/, or assets/ bundle, and the body is organized into clear sections (Usage, Instructions with numbered subsections, Examples, See also) with nothing buried. It falls short of anchor 5 only because there is no overview-to-detail split — the ~35-line Examples section and the duplicated since snippet could live in a reference file or be factored to reduce the main body. Not score 3 because structure is good and no content that clearly belongs in a separate file is inlined.

4 / 5

Total

18

/

20

Passed

Description

43%Weight 40%Scale 1-5

Based on the skill's description, can an agent find and select it at the right time? Clear, specific descriptions lead to better discovery.

The description is maximally safe for distinctiveness — an explicit-invocation-only guard that eliminates conflict risk — but it omits any statement of what the skill does. A user browsing skill descriptions learns nothing about the review-fix-iterate loop it implements. Adding a one-sentence capability summary and a few natural trigger terms would raise specificity and completeness substantially.

Suggestions

Add a 'what' clause describing concrete capabilities, e.g. "Iterative review-fix loop: reviews the current branch or commit range, fixes findings, commits, and re-reviews until all reviews pass (up to --max-iterations)."

Include natural trigger keywords such as "review and fix", "fix review findings and iterate", or "refine until reviews pass" alongside the invocation guard so the skill is discoverable by description search.

Briefly differentiate from sibling skills in the description (e.g. "unlike /roborev-fix, re-reviews after each fix") to strengthen distinctiveness beyond the invocation guard alone.

DimensionReasoningScore

Specificity

The description states only "Use only when the user explicitly invokes /roborev-refine" — no domain or capability action (review, fix, commit, re-review) is named anywhere. It avoids anchor 1's pure abstraction because it names a concrete trigger, but it fails to describe even one concrete action, matching anchor 2's minimal-action level.

2 / 5

Completeness

The 'when' is explicit ("Use only when the user explicitly invokes /roborev-refine") but the 'what' is entirely missing — nothing describes what the skill does. Anchor 2 directly covers this case: only 'when' present without 'what'. Not score 3 because a clear 'what' is a prerequisite for that anchor.

2 / 5

Trigger Term Quality

The single trigger phrase "/roborev-refine" is present and exact, but no natural-language keywords (review, fix, iterate, re-review, refinement loop) or synonyms accompany it. This matches anchor 3: some relevant keywords while missing common variations.

3 / 5

Distinctiveness Conflict Risk

The strict literal-invocation restriction ("Use only when the user explicitly invokes /roborev-refine") makes it impossible to trigger for the wrong skill and carves out a clear niche relative to sibling skills like /roborev-fix. This matches anchor 5's clear distinct triggers with minimal conflict risk.

5 / 5

Total

12

/

20

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 — 16 / 16 Passed

Validation for skill structure

No warnings or errors.

Repository
kenn-io/roborev
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.