prose-pass

Judge added comments, JSDoc, changelog entries, and PR bodies on information content and cut what a reader can derive. Use before presenting a diff, committing, or writing a PR body.

Quality

83%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

Securityby

Passed

No findings from the security scan

Quality

Content

82%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.

A well-structured, instruction-only skill body that provides concrete, non-redundant judgment criteria for trimming prose in a diff. It is lean, assumes competence, and closes with an actionable checklist and lint verification, though validation feedback loops could be made more explicit.

Suggestions

Add an explicit validate-fix-retry loop to the Checklist (e.g., 'After deletions, run lint; if it fails on a dropped JSDoc block, restore the minimal required block and re-lint') to strengthen the feedback checkpoint for this batch operation.

Consider extracting the per-artifact rules (Changelogs, PR bodies) into a one-level-deep reference file so the main body reads as the overview, improving progressive disclosure.

Tighten the overlap between 'AI smell openers' and 'Comments above a log call' since both resolve to the same narration-vs-information test, reducing redundancy.

Dimension	Reasoning	Score
Conciseness	Lean and efficient throughout; assumes Claude's competence (never explains what a diff, JSDoc, or changelog is) and every section carries a distinct, purposeful judgment rule with no padding.	5 / 5
Actionability	Concrete decision rules and tests ('Would a consumer notice this change... If not, cut it'; 'delete when the remaining sentence merely narrates the code') plus a numbered checklist, but it is judgment-based guidance rather than executable code or commands.	4 / 5
Workflow Clarity	A clear numbered Checklist sequences the pass with a re-read checkpoint and lint re-run after renames; validation is present (verify lint, re-read for obviousness) though the feedback loop is somewhat implicit rather than a hard validate-fix-retry cycle.	4 / 5
Progressive Disclosure	Well-organized with clear section headers and a single clearly-signaled one-level-deep external reference (updating-changelogs.md); no bundle files exist, so all guidance is appropriately inline with minor organization gaps from length.	4 / 5
	Total	17 / 20 Passed

Description

83%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, specific description that clearly states both the skill's purpose and its usage triggers in third person. It avoids fluff and names concrete artifact types and actions.

Dimension	Reasoning	Score
Specificity	Names the domain (comments, JSDoc, changelog entries, PR bodies) and concrete actions (judge on information content, cut what a reader can derive); lists several specific targets but not an exhaustive action set.	4 / 5
Completeness	Explicitly answers both what ('Judge added comments... and cut what a reader can derive') and when ('Use before presenting a diff, committing, or writing a PR body') with concrete trigger phrases.	5 / 5
Trigger Term Quality	Natural terms a user would say ('comments', 'JSDoc', 'changelog entries', 'PR body', 'diff', 'committing') are present with a concrete trigger ('Use before presenting a diff, committing, or writing a PR body'), though a few synonyms are missing.	4 / 5
Distinctiveness Conflict Risk	The niche (a prose-trimming pass over a diff's added text) is distinct with clear triggers; minor overlap risk with general code-review skills.	4 / 5
	Total	17 / 20 Passed

Validation

93%

Warnings & errors only

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

Criteria	Description	Result
relative_links	Relative link issues: 1 suspicious	Warning

	Total	15 / 16 Passed

Repository: Consensys-Incorporated/ocap-kernel
Path: .claude/skills/prose-pass/SKILL.md
Commit: 3d5aea9

Reviewed: 6 days ago

Table of Contents

Discovery Implementation Validation

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.