# RaceTagger API
> AI-powered sports & motorsport photo recognition. Detect race bib numbers, sponsor logos, license plates, aircraft markings. Match detections to participant data for automated photo organization.
> First API built specifically for race and sports photography analysis.

## Full Documentation
- /llms-full.txt: Complete API reference with code examples (curl, Python, JavaScript)
- /docs: User guides and API documentation hub
- /docs/getting-started: First setup and installation guide
- /docs/sport-categories: Supported sport categories and detection strategies
- /docs/participant-presets: CSV import and participant matching
- /docs/folder-organization: Automatic photo folder sorting
- /docs/metadata: IPTC/XMP metadata writing for photo files
- /docs/pro-tips: Advanced optimization techniques
- /docs/api: REST API reference

## Capabilities
- Detect bib/race numbers in sports photos (motorsport, running, cycling, motocross)
- Identify sponsor logos, team liveries, and brand text
- Read license plates (street vehicles)
- Detect aircraft registrations, airline, and model
- Visual matching: identify participants by vehicle/helmet/jersey colors when numbers aren't visible
- Match detected numbers to participant CSV data (the killer feature)
- Suggest filenames, keywords, and metadata for photo organization
- Batch processing up to 50 images per request
- Confidence scores, bounding boxes, and scene context
- Sport and subcategory auto-detection

## API Documentation
- /docs/api/openapi.yaml: Full OpenAPI 3.1 specification
- /docs/api/README.md: API overview and quickstart guide
- /docs/api/authentication.md: Authentication guide (API keys)
- /docs/api/errors.md: Error codes reference with LLM-friendly suggestions
- /docs/api/endpoints/analyze.md: Image analysis endpoint (POST /api/v1/analyze)
- /docs/api/endpoints/analyze-and-match.md: Combined analysis + participant matching (POST /api/v1/analyze-and-match)
- /docs/api/endpoints/match.md: Standalone participant matching (POST /api/v1/match — free, 0 credits)
- /docs/api/endpoints/batch.md: Batch analysis (POST /api/v1/analyze/batch)

## API Base URL
- https://racetagger.cloud/api/v1

## Authentication
- Header: Authorization: Bearer <api_key>
- Key format: rt_live_xxx (production) or rt_test_xxx (sandbox)
- Request access: https://racetagger.cloud/api-access
- Get your key: https://racetagger.cloud/dashboard/api-keys
- Sandbox tier: 100 free credits/month on approval

## Key Endpoints
- POST /api/v1/analyze — Analyze a single image (1 credit)
- POST /api/v1/analyze-and-match — Analyze + match with participant data (1 credit)
- POST /api/v1/match — Match analysis results to participants (0 credits, free)
- POST /api/v1/analyze/batch — Analyze up to 50 images (1 credit/image)
- GET /api/v1/results/:id — Retrieve analysis result by ID
- GET /api/v1/usage — Check credit balance and usage stats

## MCP Server (for AI agents)
- Remote MCP endpoint: https://racetagger.cloud/mcp
- Discovery: https://racetagger.cloud/.well-known/mcp.json
- Compatible with: Claude Desktop, Cursor, Windsurf, and any MCP client
- Tools: racetagger_analyze, racetagger_match, racetagger_search, racetagger_export, racetagger_usage

## Supported Image Formats
- JPEG, PNG, WebP (direct upload or URL)
- Base64 encoding supported
- Recommended: < 4MB per image for optimal speed

## Categories
- motorsport (default): F1, MotoGP, karting, motocross, rally, enduro
- running: marathons, trail running, cross country
- cycling: road cycling, MTB, gran fondo
- street: license plate recognition
- aviation: aircraft registration, airline, model identification

## Pricing
- Sandbox: Free, 100 credits/month
- Starter: EUR 19/month, 2000 credits
- Pro: EUR 49/month, 8000 credits
- Business: EUR 99/month, 20000 credits
- Enterprise: Custom pricing
- Match endpoint is always free (0 credits)

## Company
- Website: https://racetagger.cloud
- Support: info@racetagger.cloud
- Status: https://status.racetagger.cloud