Audit Verify Explain Grade 5
Meng ToCommunityCreated by Meng To
- Skill
- Review and audit
- Software development
- MIT
- No issues found
AutoGineer summary
Written by AutoGineer from the skill’s own files, not by the author
This skill provides a structured framework for reviewing technical work. It mandates a three-step process: auditing the source of truth, verifying claims through concrete evidence (such as tests, logs, or execution), and communicating the results in plain, grade-5 level language. It is designed to bridge the gap between complex technical implementation and clear, actionable understanding for any stakeholder.
What it does
- Audits code changes, bug fixes, and performance claims against source data.
- Prioritizes evidence-based verification using tests, logs, and direct execution.
- Translates complex technical jargon into simple, plain-language explanations.
- Distinguishes clearly between verified facts and subjective judgments.
- Provides a standardized reporting format for changes, impact, and verification status.
- Identifies unverified claims or remaining risks in the work.
When to use it
- When you need to review or audit a technical change or bug fix.
- When you must verify if a specific claim or performance improvement is accurate.
- When you need to explain technical findings to non-technical team members or stakeholders.
- When you want to ensure that reported results are backed by actual evidence rather than assumptions.
Advantages
- Reduces ambiguity by requiring concrete evidence for all claims.
- Improves communication by enforcing simple, clear language standards.
- Prevents over-claiming by explicitly stating what has and has not been verified.
- Provides a consistent, professional structure for audit reports.
Usage examples
- “Review this pull request and explain if the performance fix actually works.”
- The agent will inspect the code diff, run the relevant performance tests, and provide a report stating exactly what was measured and whether the fix is effective in simple terms.
- “Validate this bug fix and tell me what changed in plain English.”
- The agent will verify the fix by reproducing the bug, confirming the resolution, and summarizing the change and its impact without using overly technical jargon.
Install
Download the skill as a ZIP:
https://autogineer.com/agent-skills/mengto/audit-verify-explain-grade-5/downloadUnzip it into the cross-client skills folder of your project:
mkdir -p .agents/skills && unzip audit-verify-explain-grade-5.zip -d .agents/skills
Documentation by Meng To
Audit, Verify, Explain
Core Rule
Treat every answer as three jobs:
Audit what changed or what is being claimed.
Verify it with direct evidence.
Explain it like the reader is smart but new to the topic.
Do not skip verification when local files, commands, logs, tests, screenshots, or source data are available. Do not pretend something was verified if it was only inferred.
Workflow
1. Audit
Start by finding the real source of truth:
For code changes, inspect the diff, touched files, related call sites, and existing tests.
For bug fixes, identify the before/after behavior and the user-facing path.
For performance claims, separate measured evidence from likely improvement.
For release or app behavior, check the packaged/running artifact when possible.
For documents or content, compare the user request against the actual produced artifact.
Look for:
obvious bugs or regressions
missing edge cases
stale assumptions
unverified claims
mismatches between implementation and user intent
risks that a grade-5 explanation might accidentally hide
2. Verify
Prefer evidence in this order:
Automated tests, builds, linters, typechecks, or validators.
Running the actual app or workflow.
Logs, process checks, screenshots, generated artifacts, or live output.
Static code inspection when execution is impractical.
Clearly labeled inference when nothing stronger is available.
When verification fails, report the blocker and what it means. When verification is partial, say exactly what was and was not checked.
For performance work, avoid overclaiming. Say "this removes repeated work" only when the code clearly does so. Say "should improve" only when no timing trace was captured. Say "measured faster" only when before/after measurements exist.
3. Explain Simply
Use grade-5 language without talking down to the user:
Use short sentences.
Define technical terms in plain words.
Use one simple analogy only if it genuinely helps.
Say what changed, why it matters, and how to test it.
Keep important caveats visible.
Prefer this shape:
What changed:
- ...
Why it matters:
- ...
How I verified it:
- ...
What is still not proven:
- ...For very small answers, use a short paragraph instead of forcing headings.
Explanation Standards
Translate technical ideas like this:
"cache" -> "remember the answer so we do not ask the same question again"
"metadata" -> "small facts about a file, like size or modified date"
"regression" -> "something that used to work but broke"
"artifact" -> "the real file or app that was created"
"static inspection" -> "reading the code without running it"
Do not say "everything works" unless the full workflow was tested. Say "the checked parts work" when verification covered only part of the system.
Output Rules
Lead with the answer. Keep the tone calm and clear.
Include file paths, commands, commit hashes, test names, or log snippets when they are the evidence. Keep them brief.
Separate facts from judgment:
Fact: "The tests passed."
Judgment: "That gives confidence in the timeline planner, but not the full editor UI."
End with the most useful next test only when another test would materially improve confidence.
Files
- ATTRIBUTION.mdAdded by AutoGineer873 B
- LICENSELicence file added by AutoGineer1.0 KiB
- SKILL.md3.6 KiB
- agents/openai.yaml224 B
- demo/PROMPT.md2.0 KiB
- demo/expected-output.md266 B
- demo/index.html3.9 KiB
- demo/input.md272 B
- demo/preview.jpg83.0 KiB
SKILL.md
Audit, Verify, Explain
Core Rule
Treat every answer as three jobs:
Audit what changed or what is being claimed.
Verify it with direct evidence.
Explain it like the reader is smart but new to the topic.
Do not skip verification when local files, commands, logs, tests, screenshots, or source data are available. Do not pretend something was verified if it was only inferred.
Workflow
1. Audit
Start by finding the real source of truth:
For code changes, inspect the diff, touched files, related call sites, and existing tests.
For bug fixes, identify the before/after behavior and the user-facing path.
For performance claims, separate measured evidence from likely improvement.
For release or app behavior, check the packaged/running artifact when possible.
For documents or content, compare the user request against the actual produced artifact.
Look for:
obvious bugs or regressions
missing edge cases
stale assumptions
unverified claims
mismatches between implementation and user intent
risks that a grade-5 explanation might accidentally hide
2. Verify
Prefer evidence in this order:
Automated tests, builds, linters, typechecks, or validators.
Running the actual app or workflow.
Logs, process checks, screenshots, generated artifacts, or live output.
Static code inspection when execution is impractical.
Clearly labeled inference when nothing stronger is available.
When verification fails, report the blocker and what it means. When verification is partial, say exactly what was and was not checked.
For performance work, avoid overclaiming. Say "this removes repeated work" only when the code clearly does so. Say "should improve" only when no timing trace was captured. Say "measured faster" only when before/after measurements exist.
3. Explain Simply
Use grade-5 language without talking down to the user:
Use short sentences.
Define technical terms in plain words.
Use one simple analogy only if it genuinely helps.
Say what changed, why it matters, and how to test it.
Keep important caveats visible.
Prefer this shape:
What changed:
- ...
Why it matters:
- ...
How I verified it:
- ...
What is still not proven:
- ...For very small answers, use a short paragraph instead of forcing headings.
Explanation Standards
Translate technical ideas like this:
"cache" -> "remember the answer so we do not ask the same question again"
"metadata" -> "small facts about a file, like size or modified date"
"regression" -> "something that used to work but broke"
"artifact" -> "the real file or app that was created"
"static inspection" -> "reading the code without running it"
Do not say "everything works" unless the full workflow was tested. Say "the checked parts work" when verification covered only part of the system.
Output Rules
Lead with the answer. Keep the tone calm and clear.
Include file paths, commands, commit hashes, test names, or log snippets when they are the evidence. Keep them brief.
Separate facts from judgment:
Fact: "The tests passed."
Judgment: "That gives confidence in the timeline planner, but not the full editor UI."
End with the most useful next test only when another test would materially improve confidence.
Requirements and compatibility
Works with: Claude Code, Claude apps, OpenAI Codex, GitHub Copilot, Cursor, Gemini CLI, Kiro, OpenCode, Goose, Amp, Roo Code, JetBrains Junie, Factory Droid, Devin (ex-Windsurf), Mistral Vibe, OpenHands, Letta
“Works with” means the tool reads the open Agent Skills format; a skill written for one tool may use features others lack.
Security
No issues found by our static scan· No issues found
A static scan is not a guarantee. Skills are third-party content: review a skill before you let an agent use it.
Attribution and licence
- Created by
- Meng To
- Publisher
- Meng To (Community)
- Source
- MengTo/Skills · agent-skills/codex/audit-verify-explain-grade-5 · commit 5ea1661
- Licence
- MIT License (redistribution allowed)
- LICENSE (added by AutoGineer from LICENSE)
- ATTRIBUTION.md (written by AutoGineer)
- Synced
AutoGineer is not affiliated with or endorsed by Meng To or Meng To.
Download ZIP
The ZIP holds the skill’s files, its licence files and an ATTRIBUTION.md. Download audit-verify-explain-grade-5.zip