AI Agent capability routing network that discovers, inspects, and calls 10,000+ real-world tools and data sources through a unified protocol
QVeris is an AI Agent capability routing network designed to solve the fragmentation problem when agents call external tools. Through a unified Discover → Inspect → Call protocol, it enables agents to discover 10,000+ real-world capabilities via natural language, pre-inspect parameters and quality metrics before invocation, and safely execute in sandboxes with structured JSON output.
Core Routing Capabilities#
| Capability | Description | API |
|---|---|---|
| Discover | Natural language search across 10,000+ capabilities, free | POST /search |
| Inspect | Pre-view parameter schema, success rate, latency, credit cost | POST /tools/by-ids |
| Call | Sandboxed execution, structured JSON output | POST /tools/execute |
Integration Methods#
- CLI (recommended): Subprocess execution, zero prompt token consumption, for Claude Code / OpenClaw and any exec-capable agent
- MCP Server: Seamless IDE integration for Cursor / Claude Desktop and other MCP-only clients
- Python SDK: Native Python integration for Python projects and agent frameworks
- REST API: Language-agnostic for custom integrations in any language
Vertical Domain Coverage#
- Quantitative Trading & Macro Data: Index constituents, factor data, yield curves, inflation data
- Investment Research & Risk Compliance: Real-time financials, analyst consensus, valuation models, KYC, sanctions screening, beneficiary chains
- Crypto Assets & Alternative Signals: On-chain activity, DeFi protocols, stablecoin flows, news/social/event-driven signals
- General Tools: Weather, maps, search, documents, social, blockchain, healthcare, and 15+ categories
Security & Quality#
- Sandboxed execution for all calls
- Session isolation: session-level config does not modify system files
- Audit & access control: full audit logs, RBAC, rate limiting
- SLA: 99.99% availability, <500ms p95 latency
Quick Start#
Register at qveris.ai for a free API Key (includes 1,000 credits):
curl -fsSL https://qveris.ai/cli/install | bash
qveris discover "weather forecast API"
qveris inspect 1
qveris call 1 --params '{"wfo":"LWX","x":90,"y":90}'
Billing Model#
- Discover: free
- Call: 1–100 credits per invocation, priced by data and task value
- Free tier: 1,000 credits; paid $19 = 10,000 credits, never expire
Pending Confirmation#
- Whether the core routing engine is open source (README states it is a managed service, engine source not public)
- Full list of 10,000+ capabilities requires login to browse
- Enterprise VPC/private cloud options are still in planning
- Open source license type for the main repository not mentioned in available materials