iterate-until-verified

Apply a prompt-agnostic execution and verification loop to any substantial task while preserving the original request. Use when the user asks to fan out work, use subagents or independent reviewers, loop until done, benchmark against references, apply a harsh critic, compare candidates blind, improve an existing prompt with verification, or continue until explicit quality gates pass.

Quality

96%

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

100%

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

A well-crafted instruction skill: lean, actionable, with a clear gated workflow and honest stopping criteria. It earns top marks across conciseness, actionability, workflow clarity, and organization.

Dimension	Reasoning	Score
Conciseness	Lean, imperative prose with no concept-explanation padding; every section instructs rather than describes, and no tokens are spent on what Claude already knows.	3 / 3
Actionability	Provides concrete, executable guidance: a fill-in acceptance-matrix table, a numbered gate loop with pass/fail/blocked outcomes, and a copy-ready compose-mode template block.	3 / 3
Workflow Clarity	A clearly sequenced 7-step process with explicit validation checkpoints (run gates, record pass/fail/blocked, re-run failed gates) and a 'Stop honestly' terminal checklist.	3 / 3
Progressive Disclosure	A single self-contained file with well-organized sections and clear navigation; no nested or dangling references and no bundle files to misroute.	3 / 3
	Total	12 / 12 Passed

Description

92%

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, concrete description with explicit what-and-when triggers and natural user phrasings. Its only weakness is breadth: 'any substantial task' lowers distinctiveness against overlapping skills.

Suggestions

Narrow the trigger scope (e.g. 'when the user wants iterative verification with independent reviewers or gates') to reduce overlap with general-purpose execution skills.

Consider trimming the long trigger list into the most common phrasings to keep the description tighter.

Dimension	Reasoning	Score
Specificity	Lists multiple concrete actions such as 'fan out work, use subagents or independent reviewers', 'benchmark against references', 'compare candidates blind', and 'continue until explicit quality gates pass'.	3 / 3
Completeness	It states what the skill does ('execution and verification loop... while preserving the original request') and gives an explicit 'Use when...' trigger list covering both halves.	3 / 3
Trigger Term Quality	The 'Use when' clause enumerates natural phrasings a user would actually say, including 'loop until done', 'apply a harsh critic', 'compare candidates blind', and 'improve an existing prompt'.	3 / 3
Distinctiveness Conflict Risk	The verification-loop niche is distinct, but 'any substantial task' is broad enough to overlap with general-purpose execution skills; not a clean, narrowly bounded trigger set.	2 / 3
	Total	11 / 12 Passed

Validation

100%

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 — 16 / 16 Passed

Validation for skill structure

No warnings or errors.

Repository: MengTo/Skills
Path: agent-skills/codex/iterate-until-verified/SKILL.md
Commit: 93da48f

Reviewed: about 12 hours 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.