research-claim-map_skill

This skill helps you verify claims and evaluate evidence systematically by triangulating sources, rating credibility, and identifying knowledge gaps.

30

GitHub Stars

1

Bundled Files

2 months ago

Catalog Refreshed

4 months ago

First Indexed

Readme & install

Copy the install command, review bundled files from the catalogue, and read any extended description pulled from the listing source.

Installation

Preview and clipboard use veilstrat where the catalogue uses aiagentskills.

npx veilstrat add skill lyndonkl/claude --skill research-claim-map

  • SKILL.md9.1 KB

Overview

This skill creates a structured Research Claim Map to evaluate and verify specific assertions before decisions. It helps triangulate evidence, rate source credibility, identify gaps, and produce a confidence-calibrated conclusion and recommendation.

How this skill works

You give a claim or point to verify and the skill breaks it into a testable statement, collects and categorizes evidence for and against, and rates each source by evidence quality (primary/secondary/tertiary) and credibility (expertise, independence, track record, methodology). It then documents limitations, computes a numeric confidence score, and produces an actionable conclusion (believe / skeptical / reject) with next steps.

When to use it

  • Fact-check news items, social posts, or viral statistics before sharing or acting
  • Verify vendor or competitor claims during procurement or due diligence
  • Resolve conflicting reports or competing expert statements
  • Assess evidence strength for literature reviews, policy, or strategy
  • Investigate potential misinformation or unexpected data points
  • Identify knowledge gaps that require primary data collection

Best practices

  • Restate the claim precisely with dates, quantities, and scope before evaluating
  • Prioritize primary sources and independent verification over vendor material
  • Actively search for evidence against the claim to avoid confirmation bias
  • Rate credibility explicitly (expertise, independence, track record, methodology)
  • Quantify confidence (0–100%) and document key assumptions and unknowns
  • Flag when evidence is insufficient and recommend targeted follow-up steps

Example use cases

  • Assess whether a competitor truly has the customer numbers claimed in a press release
  • Triangulate conflicting statistical claims in media coverage before policy advice
  • Verify an academic finding by locating replication studies or raw datasets
  • Conduct due diligence on a vendor’s performance benchmarks cited in sales materials
  • Evaluate social media claims that could influence crisis communications

FAQ

Confidence is a calibrated synthesis of evidence quality, source credibility, and consistency across sources; it’s expressed as a numeric 0–100% score and justified with key evidence points.

What counts as primary evidence?

Primary evidence is direct observation or original records: raw datasets, transaction logs, official filings, direct measurements, or first-hand participant accounts.

Built by
VeilStrat
AI signals for GTM teams
© 2026 VeilStrat. All rights reserved.All systems operational
research-claim-map skill by lyndonkl/claude | VeilStrat