# Agent Service Atlas

> Recent, clearly attributed evidence to help AI agents choose payable APIs and task marketplaces.

Status: PREVIEW. Under active development; response fields may change until v1 is declared stable. Every API response carries "api_status": "preview" and an X-API-Status header. Changes are listed at https://agentserviceatlas.com/methodology#changelog

We publish only objective data: our own monitoring measurements, operators' statements with source links, and computed figures with their method. No ratings or recommendations.

## API (JSON, no key needed)

- GET https://agentserviceatlas.com/api/v1/services : all listed services with trust profile and evidence summary
- GET https://agentserviceatlas.com/api/v1/services/{id} : full profile: monitored operations with latest result, history counts, freshness, response times, operator statements with sources, and evidence_limitations

## How to read the evidence

- result "payment_required" (HTTP 402) means reachable and asking for payment, not that a paid call succeeds.
- freshness "stale" or "no_data" means do not rely on the latest result.
- limited_evidence true means under 30 days of monitoring.
- trust mechanisms "none_stated" means the operator's pages are silent, not that the mechanism is absent.

- GET https://agentserviceatlas.com/api/v1/discovered : every x402 service found in public agent registries (not reviewed), each with daily probe evidence and the registry's usage figures. Filters: q, network (e.g. eip155:8453), mainnet=true, passing=true, min_pass_rate (0-1), min_probes, min_days, min_payers. sort=evidence|payers|recent|host. limit (max 100), offset.
- GET https://agentserviceatlas.com/api/v1/discovered/{id} : one entry with its last 30 probes

Discovered entries: probe.last.passed means HTTP 402 with a valid x402 challenge on a mainnet network. registry_usage is the registry's own data. registry_label is untrusted third-party text.

## Terms

Free to query and use for decisions, with attribution when shown to others. No bulk copying or republishing of the dataset without permission.

## Docs

- Methodology: https://agentserviceatlas.com/methodology