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 surfmind-space/awesome-surfmind --skill fact-checkgit clone --depth 1 https://github.com/surfmind-space/awesome-surfmindWrote 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/surfmind-space/awesome-surfmind/fact-check)<a href="https://agentmods.dev/skills/surfmind-space/awesome-surfmind/fact-check"><img src="https://agentmods.dev/badge/skills/surfmind-space/awesome-surfmind/fact-check/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/surfmind-space/awesome-surfmind/fact-check"><img src="https://agentmods.dev/badge/skills/surfmind-space/awesome-surfmind/fact-check.svg" alt="Reviewed on agentmods" width="80" height="20"></a>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 | $0.00077 | $0.00592 |
| Opus 5 | $0.00039 | $0.00296 |
| Sonnet 5 | $0.00015 | $0.00118 |
| Haiku 4.5 | $0.00008 | $0.00059 |
Grade A, and why
fact-check 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 9d 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 — 37 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Fact Check
Pull the factual claims out of the page and check each against the visible context and available sources. Never invent facts, quotes, numbers, or sources, present a guess as established, or overclaim what the evidence actually supports.
- Extract concrete, checkable claims from the selected or visible content. Skip pure opinions unless they imply facts, and watch for cited sources, links, dates, and numbers that a claim should trace to.
- Check each claim against the visible context, its cited sources, and any web search or MCP/tool results available. Prefer primary and official sources when a claim affects money, health, legal, safety, or reputation decisions.
- Classify each claim: supported, weakly supported, unsupported, contradicted, or needs more evidence. If a claim is ambiguous (a date that conflates two events, a word like "all" or "always" that may overstate), split it or mark it weakly supported rather than forcing a clean verdict.
- When support is incomplete, say what is missing and what evidence would change the verdict.
Report each claim with its Status, Evidence used, Reasoning, and what to verify next. Preserve exact names, numbers, links, dates, and currencies, and don't let a single low-quality source settle a claim on its own.
Example
Page text: "The EU AI Act took effect in 2024 and bans all facial recognition."
Claim: The EU AI Act took effect in 2024. Status: Weakly supported Evidence used: The page shows a 2024 date but blurs the entry into force date with the later phased application dates. Reasoning: The Act entered into force in 2024, yet most obligations apply later, so "took effect" is only partly accurate. Verify next: The official EU legislation timeline for when each obligation starts to apply.
Claim: The EU AI Act bans all facial recognition. Status: Contradicted Evidence used: No source on the page supports a blanket ban; the linked summary lists narrow exceptions. Reasoning: The Act restricts specific real-time biometric uses rather than banning all facial recognition, so "all" overstates it. Verify next: The exact prohibited and high-risk categories in the primary text.
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.
- 9d ago First seen · 37 lines · 77 tokens per session scan A f77248930571
fact-check is a skill published in the GitHub repository surfmind-space/awesome-surfmind (5 stars, last pushed 1mo ago), licensed MIT. It adds 77 tokens to every session and 592 once invoked, about $0.0004 per session on Opus 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 2026-08-31.
Other skills, from other repositories
cs-master
Expert computer scientist for algorithms, complexity analysis, and distributed systems.
uiux-master
UI/UX expert for design systems, accessibility, and avoiding generic AI-generated design patterns.
c-master
C programming expert for memory management, systems programming, and performance optimization.
data-science
ML engineering expert for feature engineering, model debugging, and production pipelines.
go-master
Go expert for goroutines, channels, microservices, and concurrent patterns.
network-devops
Network performance expert for latency optimization, packet analysis, and infrastructure tuning.