Skip to content
Agentmart
Submit a tool

validate

Use when checking or proofreading an analysis before it's shared or acted on (a deck, a board pack, a dashboard, SQL results), including whether its…

Skill by jpenley

IndexedMIT
Write a review

Install

Works with Claude Code, Claude apps, Codex, Cursor, Gemini CLI, GitHub Copilot, VS Code, OpenCode, Goose, Windsurf. Read the repository before you give any tool access to your data.

npx agentmarket install skill/jpenley-validate

Asks which of your tools to add it to (Claude Code, Codex, Cursor, VS Code, Windsurf, Gemini CLI, Claude Desktop) and shows the change first. See https://agentmart.dev/cli.

Get the weekly diff

The weekly diff: one email a week with new agent tools, notable updates and grade changes, plus security alerts for tools you follow. Double opt-in, one-click unsubscribe.

We use your email only for this newsletter. Privacy

About

Use when checking or proofreading an analysis before it's shared or acted on (a deck, a board pack, a dashboard, SQL results), including whether its conclusions follow from the data, or when asked only to confirm that the numbers add up or the percentages are right.

Use cases: presentations, backend, data

Reviews

Write a review

No reviews yet. Installed it? Reviews come from verified installers only. How reviews work

Permissions

What this skill can do and what it asks for, from its SKILL.md. Checked Oct 25, 2026.

Tools
No tools found
Secrets
None asked for
Install scripts
None found

Network hosts it talks to 1

  • codeload.github.com · found in install command

Files it touches

  • ~/.claude/skills/validate

What's inside

Skill name
validate
Description
Use when checking or proofreading an analysis before it's shared or acted on (a deck, a board pack, a dashboard, SQL results), including whether its conclusions follow from the data, or when asked only to confirm that the numbers add up or the percentages are right.
License
MIT
Instructions
25 lines in SKILL.md · view on GitHub

Files (2)

  • SKILL.md
  • openai.yaml · agents/

SKILL.md (first lines)

# Validate

Check an analysis before anyone repeats it. The deliverable is a **verdict on every claim** the analysis makes, because a claim can be arithmetically right and still false.

## 1. List the claims

Write down each thing a reader will repeat: every headline number, every change ("up 56%"), and every cause ("the new page lifted conversion"). Recompute each from the counts and totals shown, not from the rates and percentages printed beside them.

## 2. Give the verdict

Open with one sentence answering the question that was asked. Then a table with one row per claim that doesn't hold:

| Claim | Verdict | Why |
|---|---|---|

- **Wrong**: the number itself is miscalculated.
- **Not supported**: the arithmetic is right, but the data doesn't show what the claim says (a mix shift, unequal periods, a cohort too young for its window, a changed definition, self-selection).

Each Why is one sentence carrying the recomputed figure. Claims that hold go on one line after the table.

End with one overall verdict: **ready**, **ready with the caveats listed**, or **don't present these claims as they stand**, naming the claims that block it.

When the request is narrower than this (only the arithmetic, the deck is locked, don't rewrite anything), the opening sentence answers it and the table still follows: the person presenting needs to know which claims won't survive the room. Leave the analysis itself unchanged unless asked.

From jpenley/AITuning, MIT.

More from jpenley/AITuning

Signals

Stars
0
Forks
0
Open issues
0
Last commit (this folder)
Oct 16, 2026 (13 days ago)
Last release
Oct 17, 2026 (12 days ago)

From GitHub (fetched Oct 24, 2026).

Details

License
MIT
Repository
github.com/jpenley/AITuning
First seen
Oct 24, 2026
Last checked
Oct 24, 2026
Source
GitHubgithub-graphql

More from jpenley

Cite this page

validate skill. Agent Mart. https://agentmart.dev/skills/jpenley-validate (last updated 2026-10-24)