Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx skills add ur-grue/autopunk-media-skills --skill fact-check-promptgit clone --depth 1 https://github.com/ur-grue/autopunk-media-skillsWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/ur-grue/autopunk-media-skills/fact-check-prompt)<a href="https://agentmods.dev/skills/ur-grue/autopunk-media-skills/fact-check-prompt"><img src="https://agentmods.dev/badge/skills/ur-grue/autopunk-media-skills/fact-check-prompt/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/ur-grue/autopunk-media-skills/fact-check-prompt"><img src="https://agentmods.dev/badge/skills/ur-grue/autopunk-media-skills/fact-check-prompt.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector warn
SkillSpector: 1 finding, up to medium
These are SkillSpector’s own severities. On a checked sample its high-severity flags on skills were ~96% false positives — a documented command, a public API, a “never do X” rule — so we show them as a caution to read, not a verdict. Why →
- medium Excessive Agency · line 117 Skill enables autonomous high-impact decisions without human-in-the-loop verification. Critical operations (destructive commands, financial transactions, data deletion) should require explicit user confirmation.Fix: Add human-in-the-loop confirmation for destructive, irreversible, or high-impact operations. Never auto-execute commands that modify files, send data, or alter system state.
What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00055 | $0.14532 |
| Opus 5.5 | $0.00022 | $0.05813 |
| Sonnet 5.5 | $0.00011 | $0.02906 |
| Haiku 4.5 | $0.00006 | $0.01453 |
Grade A, and why
fact-check-prompt scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 2mo ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 692 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Fact Check Prompt
What This Skill Does
Generates a systematic fact-checking checklist from a draft article, categorizing every verifiable claim by type and priority, with the specific verification question, recommended source type, and appropriate verification strategy for each. Handles six distinct claim categories — each with its own methodology, red flags, and common errors.
When To Use This Skill
- You have finished a draft and want to ensure you have not missed any checkable claim before submitting
- You are fact-checking a piece written by another reporter and need a structured inventory of everything that requires verification
- You are on deadline and need to triage — identify which claims carry the most risk and need to be checked first
- You are training a new researcher or fact-checker and want a model for what a systematic check looks like
- You are reviewing an interview-heavy piece and need to verify quotes, titles, credentials, and biographical claims
- You are checking a data-driven article and need to verify statistical claims, survey methodology, and numerical comparisons
- You are editing a piece with legal or regulatory claims and need to confirm statutory citations, court rulings, and regulatory actions
- You are reviewing a story that references scientific studies and need to verify research findings, sample sizes, and journal citations
- You are working on a historical feature and need to verify dates, sequences of events, causation claims, and archival references
What You Need To Provide
For any fact-check
Required: The draft article text (or the section to be fact-checked); the publication type (news, magazine feature, investigative, profile, opinion with reported elements, data journalism piece). Optional: Which claims you are already confident about (to exclude from the list); any specific high-risk areas you are concerned about; the publication's fact-checking standard (light editorial check vs. full verification with source documentation); whether the piece will appear in print (corrections are permanent) or online (corrections can be appended).
What ships with it
1 file beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 2mo ago First seen · 692 lines · 55 tokens per session scan A 9b708eb588c4
fact-check-prompt is a skill published in the GitHub repository ur-grue/autopunk-media-skills (46 stars, last pushed 2mo ago), licensed MIT. It adds 55 tokens to every session and 14,532 once invoked, about $0.0002 per session on Opus 5.5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-09-04.
fact-check-prompt compared with skills from other repositories
| Mod | Type | Stars | Last push | Licence | Scan | Tokens | Works with |
|---|---|---|---|---|---|---|---|
| fact-check-promptur-grue/autopunk-media-skills | Skill | 46 | 2mo ago | MIT | A | 55 | Claude Code |
| ai-hallucination-fact-check-protocolGarethManning/education-agent-skills | Skill | 977 | 10d ago | none | A | 45 | Claude Code |
| always-verify-numbers-and-linksalivirgo/Major-AI-Skills | Skill | 2 | yesterday | MIT | A | 29 | Claude Code |
| flag-speculative-statementsalivirgo/Major-AI-Skills | Skill | 2 | yesterday | MIT | A | 23 | Claude Code |
| contradictKshitijpalsinghtomar/depth-skills | Skill | 6 | 29d ago | MIT | A | 38 | Claude Code, Codex |
| phase-investigationburiedsignals/spotlight | Skill | 60 | today | none | A | 57 | Claude Code |
| phase-methodologyburiedsignals/spotlight | Skill | 60 | today | none | A | 40 | Claude Code |