CtrlK
BlogDocsLog inGet started
Tessl Logo

ce-polish

Polish a working feature through user-directed live browser feedback. Use when a functional feature needs focused UX refinement before shipping.

68

Quality

82%

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

The canonical home for this skill is ce-polish in EveryInc/compound-engineering-plugin

SKILL.md
Quality
Evals
Security

Quality

Content

86%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.

An exemplary lean instruction-only skill: a crisp overview with done/boundary criteria, a well-sequenced four-step loop, and a single well-signaled reference that verifiably owns the operational detail. The only notable gap is that in-loop failure recovery (fix not rendering, reload failure) is left implicit.

DimensionReasoningScore

Conciseness

Every sentence carries operational content ("Never work on the repository's default branch", "it never pushes or opens a PR") with zero concept explanation or padding. Lean and efficient; every token earns its place.

5 / 5

Actionability

Concrete, executable guidance throughout — "invoke `ce-commit` for the polish changes, then report the commit(s), the still-running server URL", "use a browser capability available in the active harness; if none exists, ask them to describe what they see". Deferred startup mechanics live in the bundled `references/run.md`, so only minor gaps remain in the body itself.

4 / 5

Workflow Clarity

Four clearly sequenced steps with checkpoints: explicit wait-then-iterate ordering, defined Done/exit criteria, and blockers that end the run ("A server or checkout blocker also ends the run when it is reported with the evidence needed to resume"). Not 5 because recovery loops for in-loop failures (e.g., a fix not rendering or hot reload failing) are implicit rather than spelled out, though reachability validation is delegated to the reference.

4 / 5

Progressive Disclosure

The body is a lean overview whose single reference, `references/run.md`, exists in the bundle, is clearly signaled with an explicit ownership statement ("It owns existing-worktree safety, dev-server discovery, the bundled-script calls, reachability, and the browser handoff"), and that reference clearly signals its own next tier of files one level deep. Content is appropriately split and easy to navigate.

5 / 5

Total

18

/

20

Passed

Description

78%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 concise, third-person description that explicitly answers both what and when with a reasonably natural trigger phrase. Its main weakness is abstraction: 'polish' and 'refinement' describe intent more than concrete actions, and a few common synonyms are absent.

Suggestions

Replace abstract verbs with one or two concrete capabilities, e.g., "Iterate UX fixes on a running dev server and commit them locally" — this would lift specificity without adding length.

Add a couple of natural trigger synonyms users actually say ("tweak the UI", "fine-tune", "UI polish") to broaden keyword coverage.

Sharpen the trigger clause to name the user situation concretely (e.g., "Use when the feature works but the user wants the interface polished before shipping") to further reduce overlap with general review/fix skills.

DimensionReasoningScore

Specificity

"Polish a working feature through user-directed live browser feedback" names the domain and mechanism, but the verbs stay abstract — no list of concrete actions (e.g., inspect the page, apply fixes, commit). Matches the anchor for naming the domain with 1-2 concrete actions without being comprehensive.

3 / 5

Completeness

"Polish a working feature through user-directed live browser feedback" states the what, and "Use when a functional feature needs focused UX refinement before shipping" is an explicit when with a concrete trigger phrase — both answered clearly.

5 / 5

Trigger Term Quality

"polish", "UX refinement", "working feature", and "before shipping" are phrases a user would naturally say. A few natural synonyms (tweak, fine-tune, touch up, UI polish) are missing, so it is good but not comprehensive coverage.

4 / 5

Distinctiveness Conflict Risk

"user-directed live browser feedback" carves a distinct niche versus generic review/QA skills, but "polish a feature / UX refinement" retains minor overlap risk with general code-fix or review skills.

4 / 5

Total

16

/

20

Passed

Validation

93%

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

Validation — 15 / 16 Passed

Validation for skill structure

CriteriaDescriptionResult

frontmatter_unknown_keys

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

Warning

Total

15

/

16

Passed

Repository
crdant/compound-engineering-plugin
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.