bull-detector

Whenever the user pastes in a claim, pitch, statistic, or confident assertion and wants a gut-check — a marketing page, a vendor promise, a viral 'fact', a…

Skill by aubcake

IndexedMIT★ 3

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/aubcake-bull-detector

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.

About

Whenever the user pastes in a claim, pitch, statistic, or confident assertion and wants a gut-check — a marketing page, a vendor promise, a viral 'fact', a too-clean number — run the Bull Detector: rate how much of it is bull, and point at exactly where the manure is. Use this any time something smells unsupported.

Use cases: marketing

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/bull-detector

What's inside

Skill name: bull-detector
Description: Whenever the user pastes in a claim, pitch, statistic, or confident assertion and wants a gut-check — a marketing page, a vendor promise, a viral 'fact', a too-clean number — run the Bull Detector: rate how much of it is bull, and point at exactly where the manure is. Use this any time something smells unsupported.
License: MIT
Instructions: 30 lines in SKILL.md · view on GitHub

Files (1)

SKILL.md

SKILL.md (first lines)

Application:

Use this only in conversational replies to the user. Never apply it when producing work artifacts — writing code, documents, emails, or other deliverables.

THE MECHANIC

Point the detector at the claim and watch the needle. It's tuned to unsupported confidence, weasel words, and numbers with no denominator:

```
🐂 BULL DETECTOR

   Claim:  "<the assertion under test>"

   🔔 reading: ▓▓▓▓▓▓▓░░░  HIGH BULL

   Where the manure is:
     • "<phrase>" — <why: no source / vague quantifier / unfalsifiable / cherry-picked>
     • "<phrase>" — <why>

   🌾 What WOULD make it credible:
     <the specific evidence, number, or caveat that would clear the needle>
```

Rules:
- Calibrate the needle honestly: NONE → LOW → MODERATE → HIGH → PRIZE BULL. Plenty of claims are clean — if the needle stays low, say "barely a whiff, this checks out."
- Always point at the SPECIFIC words doing the deceiving, not a vibe. "Studies show", "up to", "10x faster" (than what?), percentages with no base rate.
- Separate "this is false" from "this is unsupported" — a claim can be true and still be bull if there's no way to know it's true.
- Stay aimed at the claim, never the person who repeated it. The target is the bull, not the messenger. Then give the one question that would settle it.

From aubcake/shills, MIT.

Signals

Stars: 3
Forks: 1
Open issues: 0
Last commit (this folder): Jun 12, 2026 (4 months ago)

From GitHub (fetched Oct 24, 2026).

Details

License: MIT
Repository: github.com/aubcake/shills
First seen: Oct 24, 2026
Last checked: Oct 24, 2026
Source: GitHub github-graphql

More from aubcake

Cite this page

bull-detector skill. Agent Mart. https://agentmart.dev/skills/aubcake-bull-detector (last updated 2026-10-24)