# case.dev API Documentation for AI/LLM Consumption

## Overview
case.dev is an API platform for legal tech developers providing document processing, semantic search, 130+ AI models, and workflow automation through one unified API.

Base URL: https://api.case.dev
Documentation: https://docs.case.dev
Console: https://console.case.dev

## Authentication
All API requests require an API key in the Authorization header:
```
Authorization: Bearer YOUR_API_KEY
```

Get your API key at: https://console.case.dev

---

## Core Services

### 1. LLMs (AI Models)
Access 130+ AI models through OpenAI-compatible API
- Endpoint: POST /v1/chat/completions
- OpenAI SDK Compatible
- Models: GPT-4, GPT-5, Claude, Llama, Gemini, CaseMark Core 1
- Features: Streaming, function calling, vision, JSON mode
- Pricing: Pay-per-use with volume discounts

Example:
```bash
curl https://api.case.dev/v1/chat/completions \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "gpt-4",
    "messages": [{"role": "user", "content": "Hello"}]
  }'
```

### 2. Vaults (Document Storage + Semantic Search)
Encrypted document storage with automatic OCR and semantic search
- Create vault: POST /v1/vaults
- Upload document: POST /v1/vaults/{id}/documents
- Semantic search: POST /v1/vaults/{id}/search
- Features: 90% cost savings vs traditional vector DBs, automatic OCR, KMS encryption
- Supports: PDF, DOCX, images, up to 500MB

Example workflow:
```bash
# Create vault
curl https://api.case.dev/v1/vaults \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -d '{"name": "My Vault"}'

# Upload document
curl https://api.case.dev/v1/vaults/{vault_id}/documents \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -F "file=@document.pdf"

# Search
curl https://api.case.dev/v1/vaults/{vault_id}/search \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -d '{"query": "contract terms"}'
```

### 3. OCR (Optical Character Recognition)
Production-grade OCR for legal documents up to 500MB
- Endpoint: POST /v1/ocr
- Features: Table extraction, searchable PDFs, high accuracy
- Supports: PDF, images (JPG, PNG, TIFF)
- Output: Text, searchable PDF, structured data

Example:
```bash
curl https://api.case.dev/v1/ocr \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -F "file=@document.pdf" \
  -F "output_format=searchable_pdf"
```

### 4. Voice Transcription
High-accuracy speech-to-text
- Endpoint: POST /v1/audio/transcriptions
- Features: Speaker diarization, PII redaction, 100+ languages
- Formats: MP3, M4A, WAV, WebM
- Use cases: Depositions, court recordings, meetings

Example:
```bash
curl https://api.case.dev/v1/audio/transcriptions \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -F "file=@audio.mp3" \
  -F "model=whisper-1"
```

### 5. Workflows
700+ expert-crafted legal workflows
- List workflows: GET /v1/workflows
- Execute workflow: POST /v1/workflows/{id}/execute
- Categories: Document generation, analysis, summarization
- Server-side prompts protect your IP

Example:
```bash
curl https://api.case.dev/v1/workflows/{workflow_id}/execute \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -d '{"input": "Your document text"}'
```

### 6. Actions (Workflow Automation)
Build automation pipelines with declarative JSON
- Create action: POST /v1/actions
- Execute: POST /v1/actions/{id}/execute
- Features: Chain OCR, AI, storage with automatic retries
- Webhook notifications on completion

Example action:
```json
{
  "name": "Process Legal Document",
  "steps": [
    {"type": "ocr", "input": "document"},
    {"type": "llm", "model": "gpt-4", "prompt": "Summarize"},
    {"type": "vault", "operation": "store"}
  ]
}
```

### 7. GraphRAG (Coming Soon)
Graph-based knowledge retrieval
- Endpoint: POST /v1/vaults/{id}/graphrag/search
- Discover entity relationships and insights
- Natural language queries

### 8. FTR Convert
Convert For The Record (FTR) court recordings to M4A
- Endpoint: POST /v1/convert/ftr
- Output: Standard M4A format for transcription

### 9. Email
Send and receive emails with parsing
- Send: POST /v1/email/send
- Receive: Webhook configuration
- Features: Attachment handling, template support

---

## Common Integration Patterns

### Pattern 1: Document Processing Pipeline
1. Upload document to Vault (automatic OCR)
2. Search vault for relevant information
3. Use LLM to analyze results
4. Store analysis back in vault

### Pattern 2: Voice to Insight
1. Transcribe audio file
2. Store transcript in vault
3. Use workflow to analyze transcript
4. Generate structured output

### Pattern 3: Automated Document Analysis
1. Create Action with steps: OCR → LLM → Store
2. Trigger via API or webhook
3. Receive notification on completion
4. Retrieve results

---

## SDKs and Tools

### OpenAI SDK Compatible
Use OpenAI's official SDKs by changing the base URL:
```python
from openai import OpenAI
client = OpenAI(
    api_key="YOUR_API_KEY",
    base_url="https://api.case.dev/v1"
)
```

### Model Context Protocol (MCP)
Connect via MCP for development tools:
```
npx @case/mcp
```

---

## Key Features

1. **Developer-First**: Modern REST API, OpenAI compatibility, comprehensive docs
2. **90% Cost Savings**: Proprietary vector storage technology
3. **Enterprise Security**: KMS encryption, audit logs, RBAC, automatic PII redaction
4. **Legal-Specific**: 700+ legal workflows, legal-optimized embeddings
5. **Durable Workflows**: Automatic retries, webhook notifications, resumable execution
6. **Complete Platform**: One API for everything (no vendor stitching)

---

## Rate Limits
- Standard: 100 requests/minute
- Pro: 1000 requests/minute
- Enterprise: Custom limits

Contact sales@case.dev for enterprise plans.

---

## Error Handling
All errors return JSON with:
```json
{
  "error": {
    "message": "Error description",
    "type": "invalid_request_error",
    "code": "invalid_api_key"
  }
}
```

Common HTTP status codes:
- 200: Success
- 400: Bad Request
- 401: Unauthorized (invalid API key)
- 429: Rate limit exceeded
- 500: Server error

---

## Support
- Documentation: https://docs.case.dev
- Email: support@case.dev
- Console: https://console.case.dev

---

## Getting Started
1. Sign up at https://console.case.dev
2. Get your API key
3. Make your first request
4. Explore workflows and build

For detailed examples and guides, visit: https://docs.case.dev