A real-time fact-checking skill that validates claims against live data sources with structured verdicts and confidence scores.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install cs-verify
2. Manual Installation
Copy the skill folder to one of these locations
~/.openclaw/skills/ <project>/skills/ Priority: Workspace > Local > Bundled
3. Prompt Installation
Copy this prompt to OpenClaw to install it automatically.
Help me install cs-verify using Clawhub. If Clawhub is not installed, install it first (npm i -g clawhub).
Prefer to download?
Get the raw skill files in a ZIP archive.
What is verify-claim?
verify-claim is a sophisticated utility within the Openclaw Skills library designed to empower AI agents with real-time verification capabilities. It allows agents to cross-reference factual statements against live data repositories, ensuring that information provided to users is accurate and current. By moving beyond static datasets, this skill significantly mitigates the risk of hallucinations in LLM outputs.
The service provides highly granular feedback, including a confidence score and a freshness indicator. Whether you are building financial trackers, automated research assistants, or general-purpose fact-checkers, verify-claim provides the structural integrity and live-data access needed for reliable AI-driven communication.
verify-claim Use Cases
- Validating currency and cryptocurrency exchange rates in real-time.
- Cross-referencing geographic data such as timezones or city populations.
- Fact-checking general information using integrated APIs like Google Fact Check.
- Automating data validation for business intelligence and data entry workflows.
- Identifying trending misinformation by monitoring the most-queried claims over 24 hours.
How verify-claim Works
- The AI agent submits a claim via a structured POST request to the verification endpoint.
- Optional categorization (e.g., financial, geo, factcheck) is applied to ensure the request is routed to the most relevant data provider.
- The system queries multiple live sources or specialized APIs to gather current evidence related to the claim.
- Data is synthesized to produce a verdict such as confirmed, stale, or false, along with a confidence metric.
- The final payload is returned to the agent, allowing it to proceed with validated facts or correct its own internal state.
verify-claim Setup
To integrate this skill into your workflow, you can use the standard REST API or the provided MCP server. verify-claim is a key component for developers utilizing Openclaw Skills for robust agent performance.
API Integration
curl -X POST https://636865636b73756d.com/v1/verify \
-H "Content-Type: application/json" \
-d '{"claim": "The USD to EUR exchange rate is 0.92"}'
MCP Server Setup
npm install -g @636865636b73756d/mcp-v1
verify-claim Data Schema & Taxonomy
The verification response is returned in a structured JSON format, detailing the claim's status and source metadata. This is a standard pattern for Openclaw Skills to ensure interoperability.
| Field | Type | Description |
|---|---|---|
| verdict | string | The truth status (confirmed, stale, disputed, false, unknown) |
| confidence | number | Accuracy score between 0 and 1 |
| current_truth | string | The verified value from the data source |
| freshness | string | Indicator of how recent the source data is (e.g., live) |
| source_count | integer | Number of sources consulted during verification |
| referral_id | string | ID used for earning query credits through referrals |
verify-claim Advanced Features
- Trending Claims API: Retrieve the top 100 most-queried claims from the last 24 hours to monitor real-world information trends.
- x402 Protocol Payments: High-volume users can access the paid tier via USDC on the Base network for just $0.001 per query.
- Referral Credit System: Earn up to 20% query credits by sharing your referral_id in requests, promoting a collaborative ecosystem within Openclaw Skills.
- Agent Discovery Compliance: Fully compatible with Agent Card (.well-known/agent.json) and service metadata standards.
- Domain Routing: Specifically optimized paths for financial data, entity information, and geographic datasets for faster response times.
SKILL.md
Loading
Related Openclaw Skills
unit-convert
A comprehensive conversion utility for real-time currency exchange, physical unit transformations, and data encoding.
1-SEC Security Platform
An open-source, single-binary cybersecurity platform offering 16 specialized modules to protect Linux servers and AI agent hosts.
Agent Pipeline Skill
A standardized, multi-agent development workflow that orchestrates specialized AI agents to handle research, coding, security, and testing tasks.
Community Demand Prospecting
A comprehensive workflow for product auditing, market research, and high-intent community outreach on platforms like Reddit and X.
Data Validate
A comprehensive validation tool for AI agents to verify URL, JSON schema, email, and phone number formatting.
geocode-lookup
A precision toolkit for converting addresses to coordinates, performing reverse lookups, and calculating global distances.
Featured*
