verification-loop

This skill should be used when the user asks to "verify code", "run verification", "check quality", "validate changes", or before creating a PR. Provides comprehensive verification including build, type check, lint, tests, security scan, and diff review.

Quality

73%

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/verification-loop/SKILL.md

Quality

Content

67%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, highly actionable verification workflow with executable commands for both Python and Node stacks and explicit stop-and-fix checkpoints in the early phases. The main weaknesses are padding in low-value sections (Continuous Mode, Integration with Hooks), duplicated report template content, a dangling reference to a nonexistent examples file, and missing validation criteria in the later phases.

Suggestions

Remove the report template from the body ("Output Format") and keep it only in references/REPORT-TEMPLATE.md, pointing there instead — this alone trims ~20 lines of duplication.

Delete or substantially tighten the "Continuous Mode" and "Integration with Hooks" sections, which contain commentary and pseudo-content ("Set a mental checkpoint") rather than executable instruction.

Fix the dangling reference: examples/example-verification-report.md does not exist in the bundle — either create the example file or drop the reference.

Add explicit pass/fail criteria or stop-and-fix guidance to Phases 3 (lint), 5 (security), and 6 (diff review) to match the validation rigor of Phases 1 and 2.

Dimension	Reasoning	Score
Conciseness	The command blocks are lean and assume Claude's competence, but several sections pad the token budget: the "A comprehensive verification system" intro, the "Continuous Mode" section's "Set a mental checkpoint" pseudo-content in a markdown block, the two-line commentary-only "Integration with Hooks" section, and the report template duplicated both inline ("Output Format") and in references/REPORT-TEMPLATE.md. This fits anchor 3 (mostly efficient but includes unnecessary sections that could be tightened) rather than anchor 4's 'minor instances'.	3 / 5
Actionability	Every phase has copy-paste-ready, executable commands for both Python and Node stacks ("uv build 2>&1 \| tail -20", "npx tsc --noEmit", "pytest --cov=src --cov-report=term-missing", "pip-audit", concrete grep patterns). Minor gaps keep it at anchor 4 rather than 5: "git diff HEAD~1 --name-only" presumes exactly one commit, and the grep-based secret scan ("sk-", "api_key") is simplistic.	4 / 5
Workflow Clarity	The six phases are clearly sequenced with explicit checkpoints ("If build fails, STOP and fix before continuing"; "Fix critical ones before continuing"; a structured pass/fail report at the end). Anchor 4 rather than 5 because Phases 3, 5, and 6 lack explicit pass/fail criteria or fix-and-retry loops, leaving validation implicit there.	4 / 5
Progressive Disclosure	Good structure: a dedicated "Reference Files" section instructs "Load only what is needed" and lists one-level-deep references, and the body points to references/STACK-DETECTION.md for stack-appropriate command selection. It stays at anchor 4 rather than 5 because examples/example-verification-report.md is referenced but does not exist in the bundle, and the report template is duplicated inline instead of living solely in references/REPORT-TEMPLATE.md.	4 / 5
	Total	15 / 20 Passed

Description

80%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 that explicitly covers both what the skill does and when to use it, with concrete quoted trigger phrases and a list of six verification capabilities. Its main weaknesses are generic framing verbs ("Provides comprehensive verification"), missing common trigger synonyms, and broad terms ("check quality", "validate changes") that raise overlap risk with related code-review and testing skills.

Suggestions

Add missing natural trigger synonyms such as "run tests", "lint the code", "pre-commit check", or "CI checks" to broaden keyword coverage.

Replace the generic framing "Provides comprehensive verification including..." with direct concrete actions (e.g., "Runs build, type check, lint, tests, security scan, and diff review, then reports PASS/FAIL per gate").

Narrow the broad triggers "check quality" and "validate changes" toward the verification niche (e.g., "run all quality gates", "verify before opening a PR") to reduce conflict risk with code-review skills.

Dimension	Reasoning	Score
Specificity	"build, type check, lint, tests, security scan, and diff review" lists several concrete capability areas, but the framing "Provides comprehensive verification including..." names categories rather than concrete transitive actions. This matches anchor 4 (several specific actions, minor gaps) rather than anchor 5, whose example demonstrates multiple direct concrete actions.	4 / 5
Completeness	It explicitly answers 'when' with concrete quoted trigger phrases plus "before creating a PR", and answers 'what' with "Provides comprehensive verification including build, type check, lint, tests, security scan, and diff review" naming six concrete components. Both halves are explicit with concrete triggers, matching anchor 5 rather than anchor 4 where the 'when' could be more specific.	5 / 5
Trigger Term Quality	The quoted phrases "verify code", "run verification", "check quality", "validate changes", and "before creating a PR" are natural user utterances. Common synonyms and related phrasings ("run tests", "lint", "pre-commit", "CI checks") are missing, so it fits anchor 4 (good coverage, a few natural terms missing) rather than anchor 5's comprehensive synonym coverage.	4 / 5
Distinctiveness Conflict Risk	"check quality" and "validate changes" are broad phrases that could overlap with code-review, testing, or linting skills, and "before creating a PR" overlaps with PR-review skill triggers. This fits anchor 3 (somewhat specific but could still overlap with similar skills) rather than anchor 4, which requires only minor overlap risk with closely related skills.	3 / 5
	Total	16 / 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
frontmatter_unknown_keys	Unknown frontmatter key(s) found; consider removing or moving to metadata	Warning

	Total	15 / 16 Passed

Repository: Galaxy-Dawn/claude-scholar
Path: skills/verification-loop/SKILL.md
Commit: 9037873

Reviewed: 2 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.