TAGGUN Receipt OCR Scanning API

{
  "service_type": "platform",
  "base_url": "https://api.taggun.io",
  "auth_method": "api_key",
  "auth_config": {
    "header_name": "apikey",
    "location": "header",
    "notes": "API key is passed in the 'apikey' request header. Sign up at taggun.io to obtain a key. A demo key is sometimes available for testing."
  },
  "endpoints": [
    {
      "name": "Scan Receipt",
      "method": "POST",
      "path": "/api/v1/receipt",
      "description": "Submit an image (base64 or multipart) or PDF of a receipt and receive structured OCR/parsing output including merchant, date, total, tax, line items, and confidence scores.",
      "parameters": {
        "content_type": "image/jpeg | image/png | application/pdf (multipart or base64 body)",
        "optional": [
          "extractLineItems (boolean)",
          "extractTime (boolean)",
          "extractTax (boolean)",
          "language (ISO code)"
        ]
      }
    },
    {
      "name": "Validate Receipt",
      "method": "POST",
      "path": "/api/v1/receipt/validate",
      "description": "Submit parsed data for validation against expected fields or business rules; returns validation results."
    }
  ],
  "pricing_model": {
    "type": "freemium",
    "details": {
      "free_tier": "Limited free scanning credits on signup for evaluation",
      "paid_tiers": "Usage-based subscription tiers priced per receipt/scan volume; enterprise plans available with higher rate limits and SLA",
      "notes": "Pricing page at taggun.io lists plans; exact per-scan cost varies by volume and feature set (line-item extraction, PDF support)."
    }
  },
  "rate_limits": {
    "notes": "Rate limits depend on subscription plan. Unauthenticated or demo usage is limited; paid plans offer higher concurrency. No public, fixed global rate-limit table published — refer to account dashboard."
  },
  "capabilities": [
    "Receipt OCR and structured data extraction",
    "Support for photos, scans, and PDFs of receipts",
    "Extraction of merchant name, address, date, time, total, tax, and currency",
    "Optional line-item extraction",
    "Confidence scores per field",
    "Multiple language support",
    "REST/JSON API with API key authentication",
    "Batch or single-file processing",
    "Receipt validation endpoint"
  ],
  "raw_analysis": "TAGGUN is a receipt OCR and document data extraction service targeting fintech, expense management, accounting, and fraud-detection use cases. It provides a polished REST API that accepts receipt images (JPEG/PNG) or PDFs and returns structured JSON containing merchant details, date/time, totals, taxes, line items, and confidence scores. The service emphasizes accuracy, real-time processing, and developer-friendly integration, positioning itself as an affordable alternative to enterprise OCR suites. Maturity: Taggun has been operating since the mid-2010s and is actively used by expense apps, bookkeeping platforms, and banks; it maintains public documentation and a demo environment. Integrations: no major out-of-the-box connector marketplace is advertised; integration is primarily via direct API calls, SDKs (community-maintained), and webhook-style workflows built by consumers. Auth uses a simple API key header, lowering onboarding friction. Pricing follows a freemium/usage-based model with free evaluation credits and paid tiers scaling by scan volume and features. Because the original description is marketing-flavored and sparse, the base URL and endpoint details above are inferred from typical Taggun documentation (/api/v1/receipt); actual paths, header names, and pricing should be verified against the live docs at taggun.io. Overall, it is a mature, narrowly scoped utility API best suited for automated receipt ingestion pipelines rather than general-purpose document AI."
}

1/3 tests passed

{
  "overall": 67,
  "dimensions": {
    "token_efficiency": 9.0,
    "first_try_success": 8.0,
    "response_parseability": 9.0,
    "error_clarity": 6.5,
    "doc_quality": 7.5,
    "auth_simplicity": 7.0,
    "latency": 6.0,
    "consistency": 6.5
  },
  "pricing_normalized": {
    "model": "freemium + usage-based subscription",
    "free_tier": "limited scan credits on signup",
    "paid": "per-receipt volume tiers; enterprise SLA available",
    "transparency": "moderate — exact per-scan cost varies, not fully machine-readable"
  },
  "issues": [
    "No llms.txt or machine-readable agent onboarding file (404)",
    "No robots.txt (404) — no crawling guidance provided",
    "Free tier described only as 'limited credits', exact quota unclear",
    "Pricing varies by volume/feature set, making cost estimation hard for agents",
    "No security headers detected on homepage",
    "Homepage latency ~930ms, above ideal 500ms threshold",
    "Per-field confidence scores and multilingual support are useful, but limits/error semantics aren't explicitly documented in provided data"
  ],
  "recommendations": [
    "Publish an llms.txt with API endpoints, auth flow, and capability summary for agent consumption",
    "Expose a machine-readable pricing/quota table (JSON) with exact free-tier scan counts",
    "Add clear error-code documentation and rate-limit semantics to help first-try integration",
    "Improve homepage latency toward sub-500ms and add standard security headers",
    "Provide a public status/uptime page to support consistency perception",
    "Document exact limits of line-item extraction and PDF support availability by plan"
  ]
}