USAspending.gov

1/3 tests passed

```json
{
  "overall": 42,
  "dimensions": {
    "token_efficiency": 7.0,
    "first_try_success": 5.0,
    "response_parseability": 8.0,
    "error_clarity": 4.0,
    "doc_quality": 5.0,
    "auth_simplicity": 6.0,
    "latency": 8.5,
    "consistency": 5.0
  },
  "pricing_normalized": {
    "model": "unknown",
    "estimated_cost_per_request": null,
    "free_tier_available": false
  },
  "issues": [
    "No robots.txt or llms.txt available, making it harder for agents to discover the platform's capabilities automatically",
    "Pricing model unknown from the scan, creating uncertainty for cost-aware agents",
    "No clear getting-started or capability description detected from the initial response"
  ],
  "recommendations": [
    "Add an llms.txt file that describes API endpoints, authentication, and use cases in plain text for LLM/agent discovery",
    "Publish a clear pricing page or tier structure so agents can reason about cost before recommending",
    "Provide a well-structured API documentation page to improve first-try success for developers and agents"
  ]
}
```