An agent skill that fact-checks second-hand claims by decomposing, triaging, and verifying against primary sources.
claim-verification is easy to set up with trust notes worth reviewing. Check agent compatibility and use-case fit before adding it to your workflow.
npx skills add realnaka/claim-verificationRun the command in your terminal.
Confirm that the skill files were added to your agent workspace.
Check the README requirements before invoking the skill in your agent.
This skill teaches AI assistants to treat every second-hand claim as a rumor until proven. It breaks down long posts into individual claims, hunts down original sources, and labels each claim as confirmed, partly true, wrong, or unverified. You get a clean summary instead of a confident guess.
Claim-verification is a vendor-neutral agent skill designed to combat the spread of misinformation by AI assistants. When a user pastes a tweet, news blurb, research note, or even another AI's output, the skill activates to decompose the text into separate claims, triage them by importance, and verify each against primary sources (e.g., official filings, announcements, real data). It then labels each claim with one of four verdicts: Confirmed (✅), Partly True (🟡), Wrong/Misleading (🔴), or Unverified (⚠️). The skill is built to catch common traps like wrong attribution, echo chambers, guess-turned-fact, cherry-picking, old news as new, logo-as-contract fallacies, unsourced numbers, and false equivalences. It integrates seamlessly with Cursor and Claude Code via a single SKILL.md file, requiring no coding. Users can also configure it to run automatically on every chat. The goal is to give users a fast, honest assessment of second-hand information, helping investors, researchers, journalists, and anyone who reads the internet avoid being misled.
Looks usable, but maintenance, license, or security notes deserve a closer look.
Last commit was about 7 days ago.
0 GitHub stars indicate community interest.
0 open issues signal maintenance load.
MIT license detected.
Fact-check a viral tweet or news headline before sharing or acting on it.
Verify claims in a research note or investment thesis against official sources.
Cross-check information from another AI assistant to avoid echo chambers.
Analyze a forum post or social media thread for accuracy and completeness.
Quickly validate numbers and statistics cited in articles or presentations.
Agent skills can change coding-agent behavior; review every SKILL.md and referenced script before installing.
Check shell, network, file-system, credential, and API-key requirements before running skill workflows.
The skill relies on the AI's ability to find and interpret primary sources; accuracy depends on the underlying model and available data.
It may not catch all sophisticated disinformation or deepfakes.
Users should still apply critical thinking and not rely solely on the skill's verdicts.
0
Stars
0
Forks
0
Issues
MIT
License
A public repository of example skills for Claude, demonstrating how to create reusable instructions and resources for specialized tasks.
Adversarial AI bug hunter with auto-fix skill for Claude Code, Cursor, Codex CLI, GitHub Copilot CLI, Kiro CLI, Opencode, Pi Coding Agent, and more. Multi-agent pipeline finds security vulnerabilities, logic errors, and runtime bugs — then fixes them autonomously on a safe branch.
5 security/trust notes recorded.
Setup difficulty is 1/5.
An open-source framework to turn HTML, CSS, and animations into deterministic MP4 videos.