Reliable JSON for AI agents

Agents frequently return malformed JSON.

Deterministic inspect, safe repair, and JSON Schema validation. No LLM. No invented values.

Before: {'name': 'agent',} (single quotes, trailing comma)

After: {"name": "agent"} — only if the repair is structurally safe. Missing semantic values are never invented.

Primary: POST https://agent-json-reliability.onrender.com/v1/json/reliable

Agent Utility Store

Pay-per-call utilities for agents. Base USDC via x402. No subscription.

JSON

Web

Evidence

URL/Security

API/OpenAPI

MCP/x402

Manifest: /.well-known/agent-services.json · OpenAPI: /openapi.json · llms.txt · AGENTS.md · sitemap.xml · /.well-known/agent.json · /.well-known/x402 · /skill.md · /v1/catalog

curl

curl -s -X POST https://agent-json-reliability.onrender.com/v1/json/reliable \
  -H "content-type: application/json" \
  -d '{"text":"{'"'"'a'"'"': 1}"}'

Python

import json, os, urllib.request
BASE = os.environ.get('PUBLIC_BASE_URL', 'https://agent-json-reliability.onrender.com')
req = urllib.request.Request(
    BASE + '/v1/json/reliable',
    data=json.dumps({'text': "{'a': 1}"}).encode(),
    headers={'content-type': 'application/json'},
    method='POST',
)
print(json.load(urllib.request.urlopen(req)))

JavaScript

const BASE = process.env.PUBLIC_BASE_URL || 'https://agent-json-reliability.onrender.com';
const r = await fetch(BASE + '/v1/json/reliable', {
  method: 'POST',
  headers: {'content-type': 'application/json'},
  body: JSON.stringify({text: "{'a': 1}"}),
});
console.log(await r.json());

Optional self-reported source tags (unverified): ?source=github · ?source=x402scan · ?source=community · ?source=showcase · ?source=mcp-registry · ?source=direct