source-audit
Skill file marketing-skill/skills/source-audit/SKILL.md in alirezarezvani/claude-skills. Read the publisher’s upstream source.
Description could not be read from the source metadata.
Try your first task
A source-grounded starter task has not been verified for this skill yet.
SKILL.md
name: source-auditdescription: "Use before publishing a draft that cites links: verify that each claim is actually supported by the source it points to, link by link. Triggers include 审稿、核查信源、发布前检查、fact-check、这篇能不能发、里面说的对不对, or any review of an article, tool roundup, comparison, or review post with citations. Produces per-claim verdicts (supported, partial, contradicted, not found, unreachable) with quoted evidence. Not for copy-editing, structure review, or research from scratch."license: MITmetadata: version: 1.0.0 author: zhengqiuyang category: marketing updated: 2026-11-18Source Audit
Audit a finished draft before it is published: for every cited link, does the source actually support the claim the draft makes next to it?
Do only this. Typos, style, structure and tone belong to other skills; mixing them in hides the one failure that matters, a claim the source never made.
Before Starting
Ask only what the draft cannot answer: the draft path, the publish target (blog, newsletter, docs, social long-form; it sets how strict to be) and the scope (full audit, or only links the author flags). If a link cannot be opened from this environment, say so up front and plan to mark it unreachable.
Modes
- Full audit (default): extract, fetch, compare, write the report.
- Risk-first: only links that carry checkable facts (versions, prices, licenses, platform support, feature lists); list what was skipped.
- Single link: one verdict and its quote, no report.
Workflow
-
Extract links and context.
python3 scripts/extract_claims.py draft.md --json -o claims.jsonEach item has
url,link_textand the paragraphcontext. Iflink_countis 0 but links are visible, run--sampleto confirm the extractor works. List the checkable claims per context; mark subjective onesopinion. Read one paragraph up when context is thin, but never attribute the whole article to one link. -
Fetch every source. Never decide from a search snippet or memory; projects change. Fetch rules per source type are in references/verdict-examples.md. Never issue a verdict on an unreachable source.
-
Compare claim by claim. Every verdict needs a quoted passage that locates the evidence.
Verdict Meaning supportedThe source explicitly backs every key part of the claim. partialThe source backs part of it, or the wording is too vague to confirm. contradictedThe source explicitly says something incompatible. not-foundThe fact is absent from the source. Absence is not denial. unreachableFetch failed after a retry. opinionSubjective; not audited. Borderline cases (several links, combined facts, moved facts, summaries, vendor copy) follow the decision rules in the reference.
-
Write the report to
<draft>-audit.mdusing the reference template, and give a summary table in the reply.
Example
Draft: "Acme CLI supports Windows, macOS and Linux, and runs on 40 languages."
| Claim | Source | Evidence | Verdict |
|---|---|---|---|
| Windows, macOS and Linux | repo README | "Windows 10 2004+, macOS 12+, Ubuntu 22.04+" | supported |
| 40 languages | repo README | no language section found | not-found |
The second row is the common defect: not wrong, simply unsourced. Fix it by removing the number or citing a page that states it.
Quality Bar
- every
supportedandcontradictedverdict has a quote; - every
not-foundnames what was searched for; - every
unreachablenames the failure (status code or timeout); - opinion claims are listed as skipped, not silently removed;
- the fix list can be applied without re-reading the audit.
Anti-Patterns
- A verdict without a quote, or judged from memory or a search snippet.
- Treating absence as contradiction.
- Auditing the whole article against one link.
- Editing the draft or copy-editing inside the audit; the deliverable is a report.
Related Skills
research-summarizer: starts from sources and summarises; not for auditing a finished draft.meta-research: tests whether a hypothesis is true; not per-link verification.copywriting/md-review: use after the audit for wording and structure.