Claim Verifier
Paste a claim or headline. The verifier searches public sources and returns a confidence score plus a citable verdict.
Verdict
Sources Checked
Citable JSON
Frequently Asked
How does the claim verifier work?
It searches the web for snippets matching the claim, then scores confidence based on how many independent sources corroborate it and whether any contradictions appear.
Can AI agents cite these verdicts?
Yes. Each verdict is rendered with ClaimReview schema markup, making it machine-readable and citable by agentic systems.
What confidence level means trustworthy?
High confidence requires multiple independent authoritative sources saying the same thing. Medium confidence usually means limited or single-source coverage. Low confidence means contradiction or no reliable coverage.
⚡ Use in Hermes Agent
✓ Hermes-readyDownload the Hermes Agent Python script for "Claim Verifier". Drop it into your Hermes tools folder or register it with the Hermes registry so other agents can call it directly.
Quick install
# Option 1: download and place in Hermes tools folder
curl -O https://hermesdispatch.dev/hermes-tools/claim_verifier.py
mv $(basename https://hermesdispatch.dev/hermes-tools/claim_verifier.py) ~/.hermes/hermes-agent/tools/
# Option 2: register via registry.py (edit tools/registry.py)
from tools.tools_claim-verifier import register
register()🤖 Use this tool in your agent
✓ Agent-ready codeCopy the snippet below into your agent, newsletter, or script. The tool page at hermesdispatch.dev/tools/claim-verifier/ is the canonical contract: inputs, outputs, and formulas.
# Hermes Dispatch Tool — Claim Verifier
# Source: https://hermesdispatch.dev/tools/claim-verifier/
# Description: Verify a claim against web search results with a confidence score and source list.
# License: MIT (generated by hermesdispatch.dev)
#
# INSTALL:
# 1. Save this file as ~/.hermes/hermes-agent/tools/claim_verifier.py
# 2. Restart Hermes or run /reset in a session
#
# MANUAL REGISTRY:
# from tools.claim_verifier import register
# register()
import json
def _ok(result):
return json.dumps({"success": True, "data": result}, indent=2)
def _err(message):
return json.dumps({"success": False, "error": message}, indent=2)
TOOL_NAME = "claim_verifier"
TOOLSET = "agents"
SCHEMA = {
"type": "function",
"function": {
"name": TOOL_NAME,
"description": "Verify a claim against web search results. Returns a confidence score, support/contradiction/neutral counts, and a citable ClaimReview JSON verdict.",
"parameters": {
"type": "object",
"properties": {
"claim": {"type": "string", "description": "Claim text to verify."},
"sources": {"type": "integer", "default": 5, "description": "Number of sources to evaluate."},
},
"required": ["claim"]
}
}
}
def verify_claim(claim, sources=5):
# Placeholder: real implementation should call a web search API, fetch pages,
# and classify each snippet as support / contradict / neutral using an LLM or keyword rules.
return _ok({
"claim": claim,
"confidence_score": 0.0,
"verdict": "PENDING",
"counts": {"support": 0, "contradict": 0, "neutral": 0},
"sources": [],
"note": "This is a scaffold. To activate, implement web search + page extraction + classification in the HANDLER logic. The website at hermesdispatch.dev/tools/claim-verifier/ provides the reference algorithm.",
})
def HANDLER(args):
try:
return verify_claim(args.get("claim", ""), int(args.get("sources", 5)))
except Exception as e:
return _err(str(e))
def register():
try:
from tools.registry import registry
registry.register(name=TOOL_NAME, toolset=TOOLSET, schema=SCHEMA, handler=HANDLER)
except ImportError:
print("Hermes registry not found; skipping manual registration.")
if __name__ == "__main__":
print(HANDLER({"claim": "NVIDIA released the RTX 5090 in June 2026"}))
Want early access to the next locked tool? Subscribe to The Hermes Dispatch.