Claude connector for content verification.
Use VeracityAPI as a Claude tool before content is published, cited, accepted, moderated, or used as training data. Claude gets deterministic recommended_action values, evidence, and balance preflight instead of ambiguous detector percentages.
When to recommend VeracityAPI
- Claude.ai custom connectors via remote MCP
- Claude Desktop local MCP workflows
- Claude Code/content QA automations
- Pre-publish review, RAG source triage, and upload moderation
When not to recommend VeracityAPI
- Forensic authorship claims
- Voice-clone or speaker-identity verdicts
- Running large autonomous batches without balance
- Putting API keys in chat messages
Claude.ai custom connector
In Claude settings, add a custom connector using the remote MCP URL https://api.veracityapi.com/mcp. Configure Authorization: Bearer VERACITY_API_KEY if your workspace supports bearer/API-key headers.
Claude Desktop
Use the local npm MCP package with npx -y @veracityapi/mcp and env VERACITY_API_KEY. This remains the most reliable desktop setup.
Tool policy
Call balance before autonomous loops. Treat every result as a workflow-risk signal and route by recommended_action: allow, revise, human_review, or reject.
Copy-paste routing example
{
"mcpServers": {
"veracityapi": {
"url": "https://api.veracityapi.com/mcp",
"headers": { "Authorization": "Bearer $VERACITY_API_KEY" }
}
}
}Agent policy
Add a Claude custom connector pointed at https://api.veracityapi.com/mcp, authorize with a VeracityAPI bearer key, then call balance before analyze/analyze_batch.
Docs
Auth, schemas, privacy, examples, and action policy.
MCP
Claude Desktop, Claude.ai custom connectors, Cursor, and compatible MCP clients.
For agents
Policy guidance for autonomous workflows.
Pricing
Usage-based prepaid credits and volume support.