CtrlK
BlogDocsLog inGet started
Tessl Logo
SkillPublic

unslop-review

Rewrites code review comments so they read like a human teammate wrote them. Cuts corporate-AI throat-clearing ("I noticed...", "I was wondering if perhaps...", "It might be worth considering..."). Each comment is direct: location, the issue, a concrete fix. Use when user says...

Quality

76%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide
SecuritybySnyk

Low

Low-risk findings worth noting

Fix and improve this skill with Tessl

tessl review fix ./.agents/skills/unslop-review/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 a strong, tightly written instruction skill: concrete output format, a clear severity taxonomy, and worked bad→good examples that make the transformation unambiguous. The only real defects are the verbatim-duplicated (and truncated) 'When to Use' section and generic boilerplate in 'Limitations'.

DimensionReasoningScore

Conciseness

The body is lean and directive throughout — tight Drop/Keep lists, a one-line output grammar, and paired bad→good examples with no padding or explanations of concepts Claude already knows. Minor trims exist: the 'When to Use' section repeats the description verbatim (including its broken truncated sentence) and the 'Limitations' section is generic boilerplate. Not a 5 because of that duplicated/broken content.

4 / 5

Actionability

Fully actionable for an instruction-only skill: an exact output grammar ('L<line>: <severity prefix> <observation>. <fix>.'), a defined severity taxonomy (bug/risk/nit/q), four worked bad→good examples with concrete fixes ('Guard before user.email or early-return 404', 'wrap the call in withBackoff(3)'), and an LGTM rule. The examples are copy-paste-ready and cover the common cases.

5 / 5

Workflow Clarity

This is a simple, single-purpose skill and the single action is unambiguous: apply the Drop/Keep/Tone rules, use the Auto-clarity exceptions, emit in the given format, 'LGTM' if nothing concrete. It involves no destructive or batch operations, so the simple-skill exception applies cleanly.

5 / 5

Progressive Disclosure

No bundle files exist and none are needed; the body is organized into clear, well-labeled sections (Trigger, Format, Rules, Examples, Boundaries). It misses a 5 because the body runs ~90 lines (above the under-50-lines guideline for a 5 with sections alone) and contains trimmable boilerplate ('Limitations') plus the duplicated, truncated 'When to Use' section.

4 / 5

Total

18

/

20

Passed

Description

55%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 communicates a clear and concrete 'what', but the trigger clause is truncated mid-sentence ('Use when user says...'), which guts both the trigger term quality and the 'when' half of completeness. The single natural keyword 'code review' also creates overlap risk with genuine code-review skills.

Suggestions

Complete the truncated trigger clause with concrete phrases, e.g. 'Use when the user asks to rewrite, humanize, or de-slop PR review comments, or says the feedback sounds robotic/AI-generated.'

Add natural trigger variations such as 'PR feedback', 'humanize this comment', 'make this review sound human' so users' actual phrasings match.

Sharpen distinctiveness by explicitly scoping triggers to rewriting comment tone rather than performing code reviews, to avoid firing on plain 'code review' requests.

DimensionReasoningScore

Specificity

Names several concrete actions — 'Rewrites code review comments', 'Cuts corporate-AI throat-clearing', 'Each comment is direct: location, the issue, a concrete fix' — but coverage has minor gaps (no mention of severity prefixes or the output format). Anchor 5 requires comprehensive coverage, which this lacks; it is clearly above anchor 3's '1-2 concrete actions'.

4 / 5

Completeness

The 'what' is clear and concrete (rewrite review comments to be direct: location, issue, fix), but the 'when' is cut off mid-sentence, leaving no usable trigger guidance. Per the judging guideline, missing explicit trigger guidance caps completeness at 3; a 4 would require a complete, usable 'when' clause.

3 / 5

Trigger Term Quality

The only genuinely natural keyword is 'code review comments'; the trigger clause is literally truncated ('Use when user says...'), so intended phrases like 'humanize review' or 'make this feedback sound human' never appear. It has a relevant, natural keyword (above anchor 2's generic keywords) but is missing all common variations and synonyms.

3 / 5

Distinctiveness Conflict Risk

The tone-rewriting niche is fairly distinct, but the phrasing 'code review comments' overlaps with actual code-review skills — a user saying 'code review' likely wants a review performed, not comments rewritten. This is more than the 'minor overlap with closely related skills' of anchor 4.

3 / 5

Total

13

/

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

BigSimmo/Communication

Path

.agents/skills/unslop-review/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.