requesting-code-review

Pre-commit review: security scan, quality gates, auto-fix.

Quality

63%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

Securityby

High

Do not use without reviewing

Fix and improve this skill with Tessl

tessl review fix ./skills/software-development/requesting-code-review/SKILL.md

Quality

Content

77%

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

The content is a strong, highly actionable verification pipeline with clear sequencing and explicit feedback loops. Its main weaknesses are a small amount of redundant basic-pattern explanation and a lack of progressive disclosure that splits the long inline prompt/reference material into separate files.

Suggestions

Trim or move the 'Common Patterns to Flag' section (basic SQL injection / XSS Bad-Good examples Claude already knows) into a reference file, keeping only the project-specific scan regex inline, to improve conciseness.

Extract the verbatim delegate_task reviewer and fix-agent prompt templates into a references/ file (e.g., prompts.md) and link one level deep, shortening the SKILL.md body and improving progressive disclosure.

Keep the inline core (Steps 1-8) as the overview so the pipeline remains a single-glance workflow while the prompts and pattern reference live one level deeper.

Dimension	Reasoning	Score
Conciseness	The body is dense and operational (concrete git/grep/test commands and delegate_task calls), but the 'Reference: Common Patterns to Flag' section re-explains basic SQL injection, shell injection, and XSS with Bad/Good examples — concepts Claude already knows — so not every token earns its place.	2 / 3
Actionability	It provides fully executable, copy-paste-ready guidance: exact git diff commands, grep regex for each security check, per-language test/lint commands, complete delegate_task Python calls with full prompts, and the final commit command.	3 / 3
Workflow Clarity	Eight clearly numbered steps are sequenced with explicit validation (baseline comparison, fail-closed reviewer rules, Step 6 result evaluation) and a feedback loop in Step 7 (max 2 fix-and-reverify cycles with escalation to the user), plus a self-review checklist and a Pitfalls section.	3 / 3
Progressive Disclosure	The skill is well-sectioned and single-purpose, but at ~280 lines it keeps everything inline with no one-level-deep references; the verbatim delegate_task prompt blocks and the Common Patterns reference could be split into separate reference files to reduce body length.	2 / 3
	Total	10 / 12 Passed

Description

50%

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 concise and identifies a clear niche, but it reads as terse labels rather than a complete, trigger-rich description. It lacks an explicit 'Use when…' clause and relies on generic terms that under-specify capabilities and overlap with related skills.

Suggestions

Add an explicit 'Use when…' clause naming natural user triggers from the body (e.g., when the user says 'commit', 'push', 'ship', 'done', 'verify', or 'review before merge') to raise completeness and trigger_term_quality.

Replace the buzzword 'quality gates' with concrete actions such as 'run tests and linting, compare against a baseline' to lift specificity.

Add a short distinguishing phrase (e.g., 'verifies your own changes before committing — not for reviewing others' PRs') to reduce conflict risk with github-code-review.

Dimension	Reasoning	Score
Specificity	The description names the domain ('Pre-commit review') and three action labels ('security scan', 'quality gates', 'auto-fix'), but these are high-level category labels rather than the multiple specific concrete actions the score-3 anchor requires, and 'quality gates' is a buzzword the guidelines say to penalize.	2 / 3
Completeness	It clearly states what the skill does, but the 'when' is only implied by 'Pre-commit review' with no explicit 'Use when…' clause; per the judging guidelines, a missing explicit trigger clause caps completeness at 2.	2 / 3
Trigger Term Quality	It includes some relevant keywords a developer might say ('security scan', 'auto-fix', 'pre-commit'), but it omits the natural trigger phrasings the body itself lists ('commit', 'push', 'ship', 'verify', 'review before merge'), so coverage of natural terms is incomplete.	2 / 3
Distinctiveness Conflict Risk	The 'Pre-commit review' niche is reasonably specific, but the generic terms 'security scan' and 'quality gates' could overlap with other security/quality skills, and the description itself does not include the body's distinctness clause differentiating it from github-code-review.	2 / 3
	Total	8 / 12 Passed

Validation

81%

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 — 13 / 16 Passed

Validation for skill structure

Criteria	Description	Result
metadata_version	'metadata.version' is missing	Warning
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

	Total	13 / 16 Passed

Repository: NousResearch/hermes-agent
Path: skills/software-development/requesting-code-review/SKILL.md
Commit: 74a56b7

Reviewed: 4 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.