source-verification

source-verification is a skill for Claude Code, Codex from AtlasOmnia/hermes-custom-pack. It costs 36 tokens per session (5,120 once invoked), scanned A, a copy of source-verification, MIT.

A workflow for checking whether an article, claim, controversy, company, news item, or web page is supported by reliable evidence.

In plain words
What is it for?
It is for verifying dates, numbers, attributions, technical statements, and other factual claims in online material.
Why use it?
It separates individual claims and checks them against primary or near-primary sources instead of relying on a general summary.

Skill for Claude CodeCodex

Written for no agent in particular: nothing here depends on one.

Good fit It is for verifying dates, numbers, attributions, technical statements, and other factual claims in online material.

Compare 6 skills from other repositories ↓
Install with agentmods
npx agentmods add skills/atlasomnia/hermes-custom-pack/source-verification
Install

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.

Any agent
npx skills add AtlasOmnia/hermes-custom-pack --skill source-verification
Clone the repo
git clone --depth 1 https://github.com/AtlasOmnia/hermes-custom-pack

Made for: Claude Code, Codex.

Wrote 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.

agentmods badge for source-verification

README.md
[![agentmods](https://agentmods.dev/badge/skills/atlasomnia/hermes-custom-pack/source-verification/github.svg)](https://agentmods.dev/skills/atlasomnia/hermes-custom-pack/source-verification)
Your own site
<a href="https://agentmods.dev/skills/atlasomnia/hermes-custom-pack/source-verification"><img src="https://agentmods.dev/badge/skills/atlasomnia/hermes-custom-pack/source-verification/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.

agentmods 80×15 button for source-verification

Your own site · 80×15
<a href="https://agentmods.dev/skills/atlasomnia/hermes-custom-pack/source-verification"><img src="https://agentmods.dev/badge/skills/atlasomnia/hermes-custom-pack/source-verification.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 36 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 5,120 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 1 finding. A grade says what 26 rules found in the file — not that it is safe.
Origin 100% copy Near-identical to another mod in the catalogue.
Token cost

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.

ModelPer sessionOnce invoked
Fable 5.1 $0.00036 $0.05120
Opus 5 $0.00018 $0.02560
Sonnet 5 $0.00007 $0.01024
Haiku 4.5 $0.00004 $0.00512

Measured 8d ago against content hash 0ef7bd6f7dad, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-12, from the pricing page.

Security

Grade A, and why

source-verification scanned grade A with 1 finding 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 8d 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.

Makes network callslowCapability

Not a fault in itself. Listed so you know the mod talks to something, and to what.

- For legislature / statutes pages (for example, a state legislature site), if `web_extract` returns a giant truncated page or search results are noisy, fetch the exact section URL with `curl -L -A 'Mozilla/5.0'` and ext
Origin

This is a copy

100% identical to source-verification — 0 lines differ, which has more behind it and is treated as the original. This page carries a canonical link to it rather than competing with it.

skills/source-verification/SKILL.md · 241 lines

How it starts

The opening of the file, as written. The whole thing — 241 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Source Verification

Use this when the user asks to verify an article, claim, controversy, company/news item, or web page. The deliverable is a grounded verdict, not a generic summary.

Operating pattern

  1. Extract the article text and metadata first:
  • Title, date, author if available, publication/source, outbound links, and the exact claims being made.
  • If web_extract times out or returns empty, do not keep retrying the same path. Switch to browser automation or terminal HTTP requests with a normal User-Agent, then parse with BeautifulSoup/readability-style extraction.
  1. Break the article into checkable claims:
  • Timeline/date claims.
  • Quantitative claims: counts, stars, funding, releases, benchmark scores.
  • Attribution/provenance claims: who said what, where, when.
  • Technical claims: architecture, features, implementation details.
  1. Verify against primary or near-primary sources before relying on commentary:
  • GitHub API for repo creation dates, releases, tags, issues, licenses, stars/forks, and comment history.
  • Official docs, release notes, READMEs, technical reports, company posts.
  • Raw files from GitHub for current README/license/config evidence.
  • Search results only as discovery, not final proof.
  1. Compare source quality:
  • Distinguish first-party accusation/defense, neutral reporting, reposts/aggregators, and low-provenance blogs.
  • Flag articles with no byline, few/no outbound citations, sensational wording, or copied/reframed claims.
  1. Produce a concise verification table:
  • Claim.
  • Status: Supported / Partially supported / Unsupported / Misleading / Unverifiable.
  • Evidence.
  • Caveat.
  1. Finish with a bottom-line verdict:
  • What is solidly verified.
  • What is plausible but not proven.
  • What the article overstates.
  • What additional evidence would change the conclusion.
  1. When the user is testing whether a person is racist / antisemitic / extremist, rank evidence by strength instead of presenting a flat list:
  • Strongest: direct first-person statements, slurs, endorsements of core conspiracy claims, explicit refusals or exclusions aimed at the group, or primary-source recordings/posts.
  • Medium: repeated use of coded language or dog whistles that reputable sources trace to extremist subcultures.
  • Weakest: guilt-by-association, platform-sharing, audience overlap, or commentary about what supporters believe.
  • In the answer, lead with the strongest verified item first. If the best evidence is only medium-strength, say that plainly instead of overselling.

Read the full file on GitHub · 241 lines

Changes

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.

  1. 8d ago First seen · 241 lines · 36 tokens per session scan A 0ef7bd6f7dad

Subscribe to this mod's changes

source-verification is a skill published in the GitHub repository AtlasOmnia/hermes-custom-pack (56 stars, last pushed 28d ago), licensed MIT. It adds 36 tokens to every session and 5,120 once invoked, about $0.0002 per session on Opus 5. A static security scan graded it A with 1 finding (makes network calls). It is 100% identical to source-verification, differing in 0 lines, and is treated as a copy.