CtrlK
BlogDocsLog inGet started
Tessl Logo
SkillPublic

polish

Re-runnable pre-PR quality gate for the branch: a read-only pr-reviewer pass via the open PR, then a simplify pass that simplifies the changed code and prunes restating or ticket comments behind a test check; scope with review, simplify, or quick. Triggers on "polish my branch", "clean this up before the PR", "review and simplify", "tidy up", "prep my branch", "/polish".

Quality

90%

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

SKILL.md
Quality
Evals
Security

Quality

Content

85%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 orchestration document: lean, imperative, with concrete commands, explicit validation with revert-on-failure loops, and clearly hard safety rules. Its one real defect is bundle-level: the two files it designates as the simplify engine (rules/simplify.md, rules/quick-pass.md) are absent, leaving the most load-bearing references dangling and Pass B not fully executable from the shipped skill.

Suggestions

Ship the missing rules/simplify.md and rules/quick-pass.md in the bundle (or inline the item list and quick-pass subset), since the body references them five times as the skill's core engine and they are currently unreachable paths.

Define or point to where "the project's fast check" is specified, so the apply-then-revert-on-red loop is executable from SKILL.md alone rather than depending on the absent rules file.

Inline one concrete example of a mechanical simplify item (e.g., a restating-comment before/after) so Pass B retains some actionable guidance even when the referenced engine file cannot be loaded.

DimensionReasoningScore

Conciseness

The body is lean and imperative throughout — mode tables, exact commands, and a report template with no concept explanations (nothing teaches what git or a PR is). Even the one apparent explanation ("pr-reviewer is an agent, and Skill('pr-reviewer') errors with 'Unknown skill'") is a non-obvious operational gotcha, not padding. The only near-duplication ("Never stash, reset, or discard the user's uncommitted work" in Step 1 and Hard rules) is deliberate safety restatement, so the content matches 'every token earns its place'.

5 / 5

Actionability

Mostly copy-paste-ready: `git rev-parse --abbrev-ref HEAD`, `PR_URL=$(gh pr view --json url -q .url)`, the exact `Task(subagent_type="pr-reviewer", ...)` form, and ready commit commands. It is not fully executable though: Pass B's actual behavior ("Read and follow rules/simplify.md with items 1–9" and "the project's fast check") is delegated to files that are not present in the bundle, so the core simplify mechanics cannot be executed from what ships here — a gap beyond the anchor-5 bar of 'specific examples cover the common cases'.

4 / 5

Workflow Clarity

Steps 0–4 are clearly sequenced with explicit checkpoints and feedback loops: dirty-tree warning before committing, empty-diff and non-code-only exits, per-change validation with rollback ("each behind the project's fast check, reverting any change that turns it red"), "skip a pass with no edits", and a structured report with a Reverted list. This satisfies the batch-operation validation requirement (score would cap at 3 if missing), matching the anchor's explicit validation and error-recovery loop.

5 / 5

Progressive Disclosure

The in-body structure and signaling are good — a one-line purpose, a contents index, and clearly signaled one-level links ("[rules/simplify.md](./rules/simplify.md)", "[references/callers.md](./references/callers.md)", each with a stated purpose). However, scored against the actual bundle: rules/simplify.md and rules/quick-pass.md are referenced repeatedly as the skill's engine, yet only references/callers.md exists — the two core referenced paths are dangling, so the split content is not actually reachable. That is more than the 'minor organization gaps' of anchor 4; it fits anchor 3's 'references present but' structurally unresolved.

3 / 5

Total

17

/

20

Passed

Description

96%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.

A strong description: it states concrete capabilities in third person, explicitly answers both what it does and when to use it, and supplies a rich set of natural trigger phrases including synonyms and the slash command. The only weakness is mild overlap risk from generic triggers like 'tidy up' and 'review and simplify' against general cleanup/review skills.

DimensionReasoningScore

Specificity

Multiple concrete actions are named: "a read-only pr-reviewer pass via the open PR, then a simplify pass that simplifies the changed code and prunes restating or ticket comments behind a test check", plus explicit scoping ("scope with review, simplify, or quick"). Coverage is comprehensive of the skill's main behaviors, matching the anchor's 'multiple specific concrete actions'; not score 4 because there are no meaningful gaps in what a user needs to know to trigger and expect.

5 / 5

Completeness

Both halves are explicit: what — "Re-runnable pre-PR quality gate for the branch: a read-only pr-reviewer pass ... then a simplify pass ... prunes restating or ticket comments behind a test check"; when — "Triggers on 'polish my branch', 'clean this up before the PR', ...". This is the anchor-5 pattern of concrete what AND when with trigger phrases, so the missing-'Use when' cap of 3 does not apply.

5 / 5

Trigger Term Quality

"Triggers on 'polish my branch', 'clean this up before the PR', 'review and simplify', 'tidy up', 'prep my branch', '/polish'" provides natural user phrasings plus synonyms (polish/clean/tidy/prep) and the slash command. This matches 'comprehensive coverage of natural terms including synonyms'; not score 4 because the synonym set and explicit command form leave no common variation obviously missing.

5 / 5

Distinctiveness Conflict Risk

The pre-PR branch-gate niche is distinct and most triggers are branch/PR-anchored ("polish my branch", "clean this up before the PR", "prep my branch"), but generic phrases like "tidy up" and "review and simplify" overlap with general cleanup/review skills. This fits 'mostly distinct; minor overlap risk with closely related skills' rather than 5, whose triggers would all be unambiguous.

4 / 5

Total

19

/

20

Passed

Validation

81%

Checks the skill against the spec for correct structure and formatting. All validation checks must pass before discovery and implementation can be scored.

Validation — 13 / 16 Passed

Validation for skill structure

CriteriaDescriptionResult

metadata_field

'metadata' should map string keys to string values

Warning

frontmatter_unknown_keys

Unknown frontmatter key(s) found; consider removing or moving to metadata

Warning

relative_links

Relative link issues: 5 missing

Warning

Total

13

/

16

Passed

Repository

mthines/agent-skills

Path

skills/quality/polish/SKILL.md

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.