- Home
- Skills
- Writing & Content
- Research Claim & Citation Auditor
Works with the AI Agents you already use
Research Claim & Citation Auditor
The problem A research document can contain dozens of citations and still be poorly grounded.
Free
Research Claim & Citation Auditor
Example session with this skill installed
Audit this market report draft against the attached PDF excerpts. Focus on the claim that 'remote work adoption will double by 2026' and check if the citations actually support that timeline.
- Read your context and instructions
- Compiled the research claim
Verdict: WEAKLY GROUNDED.
Claim: Remote work will double by 2026.
Status: CONTRADICTED.
Problem: Cited source (Smith et al.) projects a 15% increase, not 100%. The 2026 date refers to the study's end-period, not a doubling event.
Rewrite: 'Remote work adoption is projected to grow by 15% through 2026.'
Connects securely to your tools. The creator never sees your data.
About this skill
The problem
A research document can contain dozens of citations and still be poorly grounded. Sources may be relevant to the topic without supporting the specific claim, statistics may be stretched beyond their original scope, and AI-generated drafts can make weak evidence sound settled.
What it does
- Breaks research into checkable claims and ranks them by importance.
- Tests whether each citation directly supports the claim attached to it.
- Flags partial support, unsupported claims, contradictions, missing citations, and unavailable sources.
- Detects citation decoration where references create authority without actually grounding the sentence.
- Checks population, timeframe, geography, causality, numerical claims, source quality, and recency.
- Separates direct source support from the writer’s own synthesis or inference.
- Produces a prioritized verification queue and rewrites claims that outrun the evidence.
Why use this skill
Most research tools focus on finding or formatting sources. This skill focuses on the harder question: does the evidence actually carry the claim being made? It makes weak support, missing provenance, and false certainty visible before the research is relied on.
Use cases
- AI-generated research briefs.
- Academic and technical writing.
- Market and competitor research.
- Literature reviews.
- Policy and due-diligence reports.
- Checking whether citations actually support an argument.
Known limitations
The skill cannot verify source content it cannot access. It never invents citations, quotations, page numbers, statistics, or study findings. Current or high-stakes claims may require external verification or qualified professional review.
Find more skills like this on the Claude Skills Marketplace, the Codex Skills Marketplace and the Gemini Skills Marketplace, or browse Claude Skills for Writing & Content.
Cannot verify truth if source text is not provided. Does not certify professional compliance for legal, medical, or financial advice. Requires manual check for paywalled sources.
How to use
- 1
Connect your agent to the Agensi MCP
Copy the prompt and paste it into Claude, ChatGPT, Cursor or any agent that supports MCP. It sets up the connection for you. Manual setup
- 2
Sign in to Agensi
Your agent shows a sign-in link. Sign in and click Allow.
- 3
Ask for what you need
Your agent loads this skill and uses it on your task. It always asks before buying a skill.
Reviews
No reviews yet
Be one of the first to try it. Every listed skill goes through the trust checks below.
Security scanned
Checked by our 8-point scan before listing
17 uses
Times used to date
Free to use
No account required to browse
Trust & safety
Security scan
Security scan passed 3 months ago
- Free with an Agensi account
Permissions
No special permissions declared or detected