Purpose
OathScore is positioned as the trust layer for AI agents, primarily serving AI trading agents and solving the problem of "every API makes promises, but who verifies them?" It consists of two products: /now and OathScore Ratings.
Main Capabilities
/now: A single endpoint that returns the current world state needed by trading agents, including exchange status, volatility, events, market regime, and data health, all obtainable in one call. The materials mention that monitoring metrics involve volatility-related indicators such as VIX/VVIX/SKEW as well as economic events.
OathScore Ratings: Independent, continuous verification of the accuracy, uptime, data freshness, and reliability of data APIs, positioned as the "credit bureau for data APIs." The materials list monitoring metric categories and include a "Rated APIs (v1)" list; however, it explicitly states that rating scores will only be populated after 30 days of monitoring data have accumulated.
How to Use
For MCP-compatible agents (such as Claude Code, Cursor, etc.): you need to first run pip install httpx mcp[cli] and clone the repository. After adding the configuration in the MCP settings (such as ~/.claude/mcp.json or Claude Desktop settings), you can ask Claude questions such as "What is the current market regime?" or "How reliable is Alpha Vantage?".
For direct HTTP calls, integration is possible via HTTP in any language.
Machine-readable discovery: Provides /llms.txt(an agent-readable product description),/llms-full.txt(complete endpoint documentation),/.well-known/ai-plugin.json(ChatGPT plugin manifest),/openapi.json(OpenAPI 3.0.3 specification), and /docs(interactive Swagger UI).
x402 micropayments: No registration required; agents pay per request using the USDC stablecoin via the x402 protocol. When rate-limited, the API returns 402 Payment Required with payment instructions.
API Audit: Provides an independent 7-day quality audit of a user's API, priced at $299–$499, requiring contact with the team.
Applicable Scenarios
Suitable for AI trading agents that need real-time market state information, as well as scenarios that require independent assessment of the quality and reliability of financial data APIs.
Limitations and Notes
Rating scores will only be populated after 30 days of monitoring data have accumulated; the project is stated to be proprietary software, with all rights reserved. The materials do not provide further details on pricing, platform coverage, the specific list of monitoring metrics, or usage results, and none are currently provided.
