Check the Claim (expanded edition): a method, prompts and a skill for checking AI claims
by PaperChaseWebb
A Markdown pack that makes your agent check income, tool, popularity, quote and funding claims before they reach a brief, a purchase or a post: arithmetic first, primary sources second, one honest verdict per claim.
- Price
- $39
- Difficulty
- Beginner
- Setup
- 10 min
- Format
- .md file
check-the-claim-expanded-edition.md · 20 KB · instant download after checkout
Want this built for you instead? We can build it for you.
What you end up with
Every claim you use carries a verdict, a labeled source for each figure, and one line on what evidence would change it.
What you get
- The six-question method and five verdicts
- Tests for five claim types
- Five arithmetic checks that break most claims before any search
- 14 prompts from intake to a finished claim card
- A drop-in claim-check SKILL.md
- Claim card, evidence log and verdict table templates
- 20 red flags, each with the follow-up question
- Five anonymized worked examples
- A credited source list
Who it is for
People who read AI income posts, tool lists and launch news and want their agent to separate what is real from what is marketing before acting on it.
Works with
Solid chips: the agent is named in this listing's own tools list. Dashed chip: Plain Markdown, so it also pastes into other agents such as Grok Bot, Hermes, OpenClaw or Muse. We list those as compatible by format only, not as tested.
Tools needed
Cost to run: Free. Uses the agent plan and web search you already have.
Time to result: First claim card in ten minutes
Sample prompts, free to use
For the claims above, run only arithmetic on the creator's own figures: per unit rate, stacked ceiling, missing denominators, count honesty, internal consistency. Show the math. Do not search the web yet. Mark any claim that fails its own arithmetic.
Most claims are settled here, before any research.
Free preview of the file
# Check the Claim (expanded edition) A PaperChaseWebb pack, version 1.0. A working method, prompts and an agent skill for checking the claims that arrive in AI content before they reach a brief, a pitch, a purchase or a post. Written for Claude, Claude Code, Codex and Grok Bot style agents. Plain Markdown, so any agent that can read a file and search the web can follow it. This pack is not a fact checking service and it gives no financial, legal or medical advice. It is the routine we rebuilt from reading dozens of AI income posts, tool lists and launch announcements in October 2026, written in our own words, with every source credited by link in `09-sources.md`. We do not repeat any creator's numbers as fact anywhere in this file. ## Files | File | What it is for | |---|---| | 01-the-method.md | The six questions every claim goes through, and the five verdicts it can land on | | 02-claim-types.md | Income claims, tool claims, star and popularity claims, quote and authority claims, funding and news claims, each with its own test | | 03-arithmetic-checks.md | The math that breaks most claims before any research: per unit rates, stacked ceilings, missing denominators | | 04-prompts.md | 14 prompts for an agent, from intake to a finished claim card | | 05-claim-check-skill.md | A drop-in SKILL.md that makes an agent run the method and refuse to upgrade a claim without evidence | | 06-templates.md | CLAIM-CARD.md, EVIDENCE-LOG.md and a verdict table you can paste into any brief | | 07-red-flags.md | 20 patterns that mean slow down, with the question each one should trigger | | 08-worked-examples.md | Five anonymized walk-throughs showing the method on real kinds of posts | | 09-sources.md | Every source we learned from, linked and credited |
Worth knowing
- This is a method, not a fact checking service. It gives no financial, legal or medical advice.
- Verdicts depend on what can be found. Some true claims will stay UNVERIFIED.
Questions
What is the difference from the free sample?
The sample has the six questions, one prompt and a short card. The full pack adds claim type tests, arithmetic checks, 14 prompts, the skill, templates, red flags and worked examples.
Does it repeat the numbers from the posts it learned from?
No. Sources are credited by link, and no creator figure is repeated as fact.
Details
Version 1.0 · Updated Oct 26, 2026 · Markdown version of this page
License: Personal and commercial use inside your own work. Do not resell or republish this pack. Download terms
Related
Check the Claim: the six questions (free sample)
Six questions, one prompt and one card template that make any agent with web search check a claim before you use it.
Vet Before You Install: ten checks, a sandbox recipe and a skill for third-party agent code
A Markdown pack for deciding whether a skill, plugin, mod, MCP server or repo is safe to let into your agent setup: read it first, pin and vendor it, trial it in a sandbox with a planted fake key, then write the decision down.
Run a daily signal board for your niche on X with an agent
A read-only agent routine and a small script that rank the last 14 days of posts in your niche by bookmarks, quotes and likes, and show what moved since yesterday.