# Case.dev

## Docs

- [Legal Agent Skills](https://docs.case.dev/agent-skills.md): Curated library of legal prompts, document templates, and research methodologies for AI agents
- [Interactive Chat](https://docs.case.dev/agents/chat.md): Persistent, multi-turn agent sessions with real-time streaming
- [Create & Configure](https://docs.case.dev/agents/create.md): Define agents with instructions, model selection, and tool restrictions
- [Agents Overview](https://docs.case.dev/agents/index.md): Run AI agents in sandboxed environments with full tool access
- [Monitor & Analyze](https://docs.case.dev/agents/monitoring.md): Poll status, get full audit trails, and register webhooks
- [Execute Runs](https://docs.case.dev/agents/runs.md): Create runs, execute them, and understand the lifecycle
- [Sandbox Environment](https://docs.case.dev/agents/sandbox.md): What's inside the agent sandbox — tools, CLI, and capabilities
- [Cancel active chat generation](https://docs.case.dev/api-reference/agents/cancel-active-chat-generation.md): Aborts the active OpenCode generation for this chat session.
- [Cancel run](https://docs.case.dev/api-reference/agents/cancel-run.md): Cancels a running or queued run. Idempotent — cancelling a finished run returns its current status.
- [Create agent](https://docs.case.dev/api-reference/agents/create-agent.md): Creates a new agent definition with a scoped API key. The agent can then be used to create and execute runs.
- [Create interactive chat session](https://docs.case.dev/api-reference/agents/create-interactive-chat-session.md): Creates a persistent OpenCode chat session in a Modal sandbox. Session state is retained and can be resumed across requests.
- [Create run](https://docs.case.dev/api-reference/agents/create-run.md): Creates a run in queued state. Call POST /agent/v1/run/:id/exec to start execution.
- [Delete agent](https://docs.case.dev/api-reference/agents/delete-agent.md): Soft-deletes an agent and revokes its scoped API key.
- [End chat session](https://docs.case.dev/api-reference/agents/end-chat-session.md): Snapshots and terminates the active sandbox (if any), then marks the chat as ended.
- [Execute ad-hoc agent](https://docs.case.dev/api-reference/agents/execute-ad-hoc-agent.md): Creates an ephemeral agent and immediately executes a run. Returns the run ID for polling status and results. This is the fastest way to run an agent without managing agent lifecycle.
- [Execute run](https://docs.case.dev/api-reference/agents/execute-run.md): Starts execution of a queued run. The agent runs in a durable workflow — poll /run/:id/status for progress.
- [Get agent](https://docs.case.dev/api-reference/agents/get-agent.md): Retrieves a single agent definition by ID.
- [Get run details](https://docs.case.dev/api-reference/agents/get-run-details.md): Full audit trail for a run including output, steps (tool calls, text), and token usage.
- [Get run status](https://docs.case.dev/api-reference/agents/get-run-status.md): Lightweight status poll for a run. Use /run/:id/details for the full audit trail.
- [List agents](https://docs.case.dev/api-reference/agents/list-agents.md): Lists all active agents for the authenticated organization.
- [Respond to chat message](https://docs.case.dev/api-reference/agents/respond-to-chat-message.md): Streams a single assistant turn as normalized state events with stable turn, message, and part ids.
- [Send chat message](https://docs.case.dev/api-reference/agents/send-chat-message.md): Proxies a message to the OpenCode session bound to this chat.
- [Stream chat events](https://docs.case.dev/api-reference/agents/stream-chat-events.md): Relays OpenCode SSE events for this chat. Supports replay from buffered events using Last-Event-ID.
- [Stream run events](https://docs.case.dev/api-reference/agents/stream-run-events.md): Streams real-time run events over SSE. Supports replay using Last-Event-ID.
- [Update agent](https://docs.case.dev/api-reference/agents/update-agent.md): Updates an agent definition. Only provided fields are changed.
- [Watch run](https://docs.case.dev/api-reference/agents/watch-run.md): Register a callback URL to receive notifications when the run completes. URL must use https and must not point to a private network.
- [Cancel deployment](https://docs.case.dev/api-reference/applications-deployments/cancel-deployment.md): Cancel a running deployment
- [Create deployment](https://docs.case.dev/api-reference/applications-deployments/create-deployment.md): Trigger a new deployment for a project
- [Create deployment](https://docs.case.dev/api-reference/applications-deployments/create-deployment-1.md): Trigger a new deployment for a project.
- [Deploy from files](https://docs.case.dev/api-reference/applications-deployments/deploy-from-files.md): Create a deployment from raw file contents (for Thurgood sandbox deployments)
- [Get deployment](https://docs.case.dev/api-reference/applications-deployments/get-deployment.md): Get details of a specific deployment including build logs
- [Get deployment logs](https://docs.case.dev/api-reference/applications-deployments/get-deployment-logs.md): Get build logs for a specific deployment
- [Get deployment status](https://docs.case.dev/api-reference/applications-deployments/get-deployment-status.md): Get the current status of a deployment
- [Get workflow status](https://docs.case.dev/api-reference/applications-deployments/get-workflow-status.md): Get current deployment workflow status and accumulated events
- [List deployments](https://docs.case.dev/api-reference/applications-deployments/list-deployments.md): List deployments for a project
- [List project deployments](https://docs.case.dev/api-reference/applications-deployments/list-project-deployments.md): List deployments for a specific project
- [Stream deployment progress](https://docs.case.dev/api-reference/applications-deployments/stream-deployment-progress.md): Stream real-time deployment progress events via Server-Sent Events
- [Add domain](https://docs.case.dev/api-reference/applications-domains/add-domain.md): Add a custom domain to a project
- [List domains](https://docs.case.dev/api-reference/applications-domains/list-domains.md): List all domains configured for a project
- [Remove domain](https://docs.case.dev/api-reference/applications-domains/remove-domain.md): Remove a domain from a project
- [Create environment variable](https://docs.case.dev/api-reference/applications-env-vars/create-environment-variable.md): Create a new environment variable for a project
- [Delete environment variable](https://docs.case.dev/api-reference/applications-env-vars/delete-environment-variable.md): Delete an environment variable from a project
- [List environment variables](https://docs.case.dev/api-reference/applications-env-vars/list-environment-variables.md): List all environment variables for a project (values are hidden unless decrypt=true)
- [Create project](https://docs.case.dev/api-reference/applications-projects/create-project.md): Create a new web application project
- [Delete project](https://docs.case.dev/api-reference/applications-projects/delete-project.md): Delete a web application project
- [Get project](https://docs.case.dev/api-reference/applications-projects/get-project.md): Get details of a specific web application project
- [Get runtime logs](https://docs.case.dev/api-reference/applications-projects/get-runtime-logs.md): Get runtime/function logs for a project
- [List projects](https://docs.case.dev/api-reference/applications-projects/list-projects.md): List all web application projects
- [Create compute environment](https://docs.case.dev/api-reference/compute/create-compute-environment.md): Creates a new compute environment for running serverless workloads. Each environment gets its own isolated namespace with a unique domain for hosting applications and APIs. The first environment created becomes the default environment for the organization.
- [Create secret group](https://docs.case.dev/api-reference/compute/create-secret-group.md): Creates a new secret group in a compute environment. Secret groups organize related secrets for use in serverless functions and workflows. If no environment is specified, the group is created in the default environment.

**Features:**
- Organize secrets by logical groups (e.g., database, APIs, third-party services)
- Environment-based isolation
- Validation of group names
- Conflict detection for existing groups
- [Delete compute environment](https://docs.case.dev/api-reference/compute/delete-compute-environment.md): Permanently delete a compute environment and all its associated resources. This will stop all running deployments and clean up related configurations. The default environment cannot be deleted if other environments exist.
- [Delete secret group or key](https://docs.case.dev/api-reference/compute/delete-secret-group-or-key.md): Delete an entire secret group or a specific key within a secret group. When deleting a specific key, the remaining secrets in the group are preserved. When deleting the entire group, all secrets and the group itself are removed.
- [Get compute environment](https://docs.case.dev/api-reference/compute/get-compute-environment.md): Retrieve a specific compute environment by name. Returns environment configuration including status, domain, and metadata for your serverless compute infrastructure.
- [Get compute pricing information](https://docs.case.dev/api-reference/compute/get-compute-pricing-information.md): Returns current pricing for GPU instances. Prices are fetched in real-time and include a 20% platform fee. For detailed instance types and availability, use GET /compute/v1/instance-types.
- [Get compute usage and billing](https://docs.case.dev/api-reference/compute/get-compute-usage-and-billing.md): Returns detailed compute usage statistics and billing information for your organization. Includes GPU and CPU hours, total runs, costs, and breakdowns by environment. Use optional query parameters to filter by specific year and month.
- [Get GPU instance](https://docs.case.dev/api-reference/compute/get-gpu-instance.md): Retrieves detailed information about a GPU instance including SSH connection details, vault mount scripts, real-time cost tracking, and current status. SSH private key included for secure access.
- [Health check](https://docs.case.dev/api-reference/compute/health-check.md): Unauthenticated health check for the Compute service.
- [Launch GPU instance](https://docs.case.dev/api-reference/compute/launch-gpu-instance.md): Launches a new GPU compute instance with automatic SSH key generation. Supports mounting Case.dev Vaults as filesystems and configurable auto-shutdown. Instance boots in ~2-5 minutes. Perfect for batch OCR processing, AI model training, and intensive document analysis workloads.
- [List environments](https://docs.case.dev/api-reference/compute/list-environments.md): Retrieve all compute environments for your organization. Environments provide isolated execution contexts for running code and workflows.
- [List GPU instance types](https://docs.case.dev/api-reference/compute/list-gpu-instance-types.md): Retrieves all available GPU instance types with pricing, specifications, and regional availability. Includes T4, A10, A100, H100, and H200 GPUs powered by Lambda Labs. Perfect for AI model training, inference workloads, and legal document OCR processing at scale.
- [List GPU instances](https://docs.case.dev/api-reference/compute/list-gpu-instances.md): Retrieves all GPU compute instances for your organization with real-time status updates from Lambda Labs. Includes pricing, runtime metrics, and auto-shutdown configuration. Perfect for monitoring AI workloads, document processing jobs, and cost tracking.
- [List secret groups](https://docs.case.dev/api-reference/compute/list-secret-groups.md): Retrieve all secret groups for a compute environment. Secret groups organize related secrets (API keys, credentials, etc.) that can be securely accessed by compute jobs during execution.
- [List secret keys in a group](https://docs.case.dev/api-reference/compute/list-secret-keys-in-a-group.md): Retrieve the keys (names) of secrets in a specified group within a compute environment. For security reasons, actual secret values are not returned - only the keys and metadata.
- [Set default environment](https://docs.case.dev/api-reference/compute/set-default-environment.md): Sets a compute environment as the default for the organization. Only one environment can be default at a time - setting a new default will automatically unset the previous one.
- [Set secrets in group](https://docs.case.dev/api-reference/compute/set-secrets-in-group.md): Set or update secrets in a compute secret group. Secrets are encrypted with AES-256-GCM. Use this to manage environment variables and API keys for your compute workloads.
- [Terminate GPU instance](https://docs.case.dev/api-reference/compute/terminate-gpu-instance.md): Terminates a running GPU instance, calculates final cost, and cleans up SSH keys. This action is permanent and cannot be undone. All data on the instance will be lost.
- [Create database branch](https://docs.case.dev/api-reference/database/create-database-branch.md): Creates a new branch from the specified parent branch (or default 'main' branch). Branches provide instant point-in-time clones of your database for isolated development, staging, testing, or feature work. Perfect for testing schema changes, running migrations safely, or creating ephemeral preview environments.
- [Create database project](https://docs.case.dev/api-reference/database/create-database-project.md): Creates a new serverless Postgres database project powered by Neon. Includes automatic scaling, connection pooling, and a default 'main' branch with 'neondb' database. Supports branching for isolated dev/staging environments. Perfect for case management applications, document workflows, and litigation support systems.
- [Database service health check](https://docs.case.dev/api-reference/database/database-service-health-check.md): Check the health status of the Database service. Returns service availability and timestamp for monitoring purposes.
- [Delete database project](https://docs.case.dev/api-reference/database/delete-database-project.md): Permanently deletes a database project from Neon and marks it as deleted in Case.dev. This action cannot be undone and will destroy all data including branches and databases. Use with caution.
- [Get connection string](https://docs.case.dev/api-reference/database/get-connection-string.md): Retrieves the PostgreSQL connection URI for a database project. Supports selecting specific branches and pooled vs direct connections. Connection strings include credentials and should be stored securely. Use for configuring applications and deployment environments.
- [Get database project](https://docs.case.dev/api-reference/database/get-database-project.md): Retrieves detailed information about a specific database project including branches, databases, storage/compute metrics, connection host, and linked deployments. Fetches live usage statistics from Neon API.
- [Get database usage and billing](https://docs.case.dev/api-reference/database/get-database-usage-and-billing.md): Returns detailed database usage statistics and billing information for the current billing period. Includes compute hours, storage, data transfer, and branch counts with associated costs broken down by project.
- [List database branches](https://docs.case.dev/api-reference/database/list-database-branches.md): Retrieves all branches for a database project. Branches enable isolated development and testing environments with instant point-in-time cloning. Each branch includes the default branch and any custom branches created for staging, testing, or feature development.
- [List database projects](https://docs.case.dev/api-reference/database/list-database-projects.md): Retrieves all serverless Postgres database projects for the authenticated organization. Includes storage and compute metrics, plus linked deployments from Thurgood apps and Compute instances.
- [Create format template](https://docs.case.dev/api-reference/format/create-format-template.md): Create a new format template for document formatting. Templates support variables using `{{variable}}` syntax and can be used for captions, signatures, letterheads, certificates, footers, or custom formatting needs.
- [Format documents with professional layouts](https://docs.case.dev/api-reference/format/format-documents-with-professional-layouts.md): Convert Markdown, JSON, or text content to professionally formatted PDF, DOCX, or HTML documents. Supports template components with variable interpolation for creating consistent legal documents like contracts, briefs, and reports.
- [Get format template](https://docs.case.dev/api-reference/format/get-format-template.md): Retrieve a specific document format template by ID. Format templates define how documents should be structured and formatted for specific legal use cases such as contracts, briefs, or pleadings.
- [Health check](https://docs.case.dev/api-reference/format/health-check.md): Unauthenticated health check for the Format service.
- [List format templates](https://docs.case.dev/api-reference/format/list-format-templates.md): Retrieve all format templates for the organization. Templates define reusable document formatting patterns with customizable variables for consistent legal document generation.

Filter by type to get specific template categories like contracts, pleadings, or correspondence.
- [Deep legal research](https://docs.case.dev/api-reference/legal/deep-legal-research.md): Perform comprehensive legal research with multiple query variations. Uses advanced deep search to find relevant sources across different phrasings of the legal issue.
- [Extract citations from URL](https://docs.case.dev/api-reference/legal/extract-citations-from-url.md): Extract all legal citations and references from a document URL. Returns structured citation data including case citations, statute references, and regulatory citations.
- [Find legal sources](https://docs.case.dev/api-reference/legal/find-legal-sources.md): Search for legal sources including cases, statutes, and regulations from authoritative legal databases. Returns ranked candidates. Always verify with legal.verify() before citing.
- [Find similar legal sources](https://docs.case.dev/api-reference/legal/find-similar-legal-sources.md): Find cases and documents similar to a given legal source. Useful for finding citing cases, related precedents, or similar statutes.
- [Get full text of legal document](https://docs.case.dev/api-reference/legal/get-full-text-of-legal-document.md): Retrieve the full text content of a legal document. Use after verifying the source with legal.verify(). Returns complete text with optional highlights and AI summary.
- [Health check](https://docs.case.dev/api-reference/legal/health-check.md): Unauthenticated health check for the Citations service. Verifies connectivity to the CourtListener API.
- [List CourtListener courts](https://docs.case.dev/api-reference/legal/list-courtlistener-courts.md): Returns CourtListener court IDs and names for docket filtering. Use these IDs in legal.docket() as the court parameter.
- [Look up USPTO trademark status](https://docs.case.dev/api-reference/legal/look-up-uspto-trademark-status.md): Look up trademark status and details from the USPTO Trademark Status & Document Retrieval (TSDR) system. Supports lookup by serial number or registration number. Returns mark text, status, owner, goods/services, Nice classification, filing/registration dates, and more.
- [Parse citations from text](https://docs.case.dev/api-reference/legal/parse-citations-from-text.md): Parses legal citations from text and returns structured Bluebook components (case name, reporter, volume, page, year, court). Accepts either a single citation or a full text block.
- [Resolve jurisdiction](https://docs.case.dev/api-reference/legal/resolve-jurisdiction.md): Search for a jurisdiction by name. Returns matching jurisdictions with their IDs for use in legal.find() and other legal research endpoints.
- [Search federal dockets and filings](https://docs.case.dev/api-reference/legal/search-federal-dockets-and-filings.md): Search federal court dockets or retrieve a specific docket with optional filing entries via CourtListener RECAP data.
- [Search USPTO patent applications](https://docs.case.dev/api-reference/legal/search-uspto-patent-applications.md): Search the USPTO Open Data Portal for US patent applications and granted patents. Supports free-text queries, field-specific search, filters by assignee/inventor/status/type, date ranges, and pagination. Covers applications filed on or after January 1, 2001. Data is refreshed daily.
- [Verify legal citations](https://docs.case.dev/api-reference/legal/verify-legal-citations.md): Validates legal citations against authoritative case law sources (CourtListener database of ~10M cases). Returns verification status and case metadata for each citation found in the input text. Accepts either a single citation or a full text block containing multiple citations.
- [Chat completions](https://docs.case.dev/api-reference/llms/chat-completions.md): Create a completion for the provided prompt and parameters. Compatible with OpenAI's chat completions API. Supports 40+ models including GPT-4, Claude, Gemini, and CaseMark legal AI models. Includes streaming support, token counting, and usage tracking.
- [Create embeddings](https://docs.case.dev/api-reference/llms/create-embeddings.md): Create vector embeddings from text using OpenAI-compatible models. Perfect for semantic search, document similarity, and building RAG systems for legal documents.
- [Get LLM configuration](https://docs.case.dev/api-reference/llms/get-llm-configuration.md): Retrieves the AI Gateway configuration including all available language models and their specifications. This endpoint returns model information compatible with the Vercel AI SDK Gateway format, making it easy to integrate with existing AI applications.

Use this endpoint to:
- Discover available language models
- Get model specifications and pricing
- Configure AI SDK clients
- Build model selection interfaces
- [Health check](https://docs.case.dev/api-reference/llms/health-check.md): Unauthenticated health check for the LLM service.
- [List available language models](https://docs.case.dev/api-reference/llms/list-available-language-models.md): Retrieve a list of all available language models from 40+ providers including OpenAI, Anthropic, Google, and Case.dev's specialized legal models. Returns OpenAI-compatible model metadata with pricing information.

This endpoint is compatible with OpenAI's models API format, making it easy to integrate with existing applications.
- [Add memory](https://docs.case.dev/api-reference/memory/add-memory.md): Store memories from conversation messages. Automatically extracts facts and handles deduplication.

Use tag_1 through tag_12 for filtering - these are generic indexed fields you can use for any purpose:
- Legal app: tag_1=client_id, tag_2=matter_id
- Healthcare: tag_1=patient_id, tag_2=encounter_id
- E-commerce: tag_1=customer_id, tag_2=order_id
- [Bulk delete memories](https://docs.case.dev/api-reference/memory/bulk-delete-memories.md): Delete multiple memories matching tag filter criteria. CAUTION: This will delete all matching memories for your organization.
- [Delete memory](https://docs.case.dev/api-reference/memory/delete-memory.md): Delete a single memory by its ID.
- [Get memory](https://docs.case.dev/api-reference/memory/get-memory.md): Retrieve a single memory by its ID.
- [Health check](https://docs.case.dev/api-reference/memory/health-check.md): Check if the Memory service is available.
- [List memories](https://docs.case.dev/api-reference/memory/list-memories.md): List all memories with optional filtering by tags and category.
- [Search memories](https://docs.case.dev/api-reference/memory/search-memories.md): Search memories using semantic similarity. Filter by tag fields to narrow results.

Use tag_1 through tag_12 for filtering - these are generic indexed fields you define:
- Legal app: tag_1=client_id, tag_2=matter_id
- Healthcare: tag_1=patient_id, tag_2=encounter_id
- [Download OCR results](https://docs.case.dev/api-reference/ocr/download-ocr-results.md): Download OCR processing results in various formats. Returns the processed document as text extraction, structured JSON with coordinates, searchable PDF with text layer, or the original uploaded document.
- [Get OCR job status](https://docs.case.dev/api-reference/ocr/get-ocr-job-status.md): Retrieve the status and results of an OCR job. Returns job progress, extracted text, and metadata when processing is complete.
- [Health check](https://docs.case.dev/api-reference/ocr/health-check.md): Unauthenticated health check for the OCR service.
- [Process document with OCR](https://docs.case.dev/api-reference/ocr/process-document-with-ocr.md): Submit a document for OCR processing to extract text, detect tables, forms, and other features. Supports PDFs, images, and scanned documents. Returns a job ID that can be used to track processing status.
- [Detect privileged content](https://docs.case.dev/api-reference/privilege/detect-privileged-content.md): Analyzes text or vault documents for legal privilege. Detects attorney-client privilege, work product doctrine, common interest privilege, and litigation hold materials.

Returns structured privilege flags with confidence scores and policy-friendly rationale suitable for discovery workflows and privilege logs.

**Size Limit:** Maximum 200,000 characters (larger documents rejected).

**Permissions:** Requires `chat` permission. When using `document_id`, also requires `vault` permission.

**Note:** When analyzing vault documents, results are automatically stored in the document's `privilege_analysis` metadata field.
- [Comprehensive multi-step research](https://docs.case.dev/api-reference/search/comprehensive-multi-step-research.md): Performs deep research by conducting multi-step analysis, gathering information from multiple sources, and providing comprehensive insights. Ideal for legal research, case analysis, and due diligence investigations.
- [Extract content from URLs](https://docs.case.dev/api-reference/search/extract-content-from-urls.md): Scrapes and extracts text content from web pages, PDFs, and documents. Useful for legal research, evidence collection, and document analysis. Supports live crawling, subpage extraction, and content summarization.
- [Find similar content](https://docs.case.dev/api-reference/search/find-similar-content.md): Find web pages and documents similar to a given URL. Useful for legal research to discover related case law, statutes, or legal commentary that shares similar themes or content structure.
- [Generate AI-powered answers from web search](https://docs.case.dev/api-reference/search/generate-ai-powered-answers-from-web-search.md): Generate comprehensive answers to questions using web search results. Supports two modes: native provider answers or custom LLM-powered answers using Case.dev's AI gateway. Perfect for legal research, fact-checking, and gathering supporting evidence for cases.
- [Get research status and results](https://docs.case.dev/api-reference/search/get-research-status-and-results.md): Retrieve the status and results of a deep research task by ID. Supports both standard JSON responses and streaming for real-time updates as the research progresses. Research tasks analyze topics comprehensively using web search and AI synthesis.
- [Health check](https://docs.case.dev/api-reference/search/health-check.md): Unauthenticated health check for the Search service.
- [Perform intelligent web search](https://docs.case.dev/api-reference/search/perform-intelligent-web-search.md): Executes intelligent web search queries with advanced filtering and customization options. Ideal for legal research, case law discovery, and gathering supporting documentation for litigation or compliance matters.
- [Read skill](https://docs.case.dev/api-reference/skills/read-skill.md): Read the full content of a legal skill by its slug. Returns markdown content, tags, and metadata.
- [Search skills](https://docs.case.dev/api-reference/skills/search-skills.md): Search the Legal Skills Store using hybrid search (text + tag + semantic). Returns ranked results with relevance scores.
- [Convert documents between formats](https://docs.case.dev/api-reference/superdoc/convert-documents-between-formats.md): Convert documents between formats using SuperDoc. Supports DOCX to PDF, Markdown to DOCX, and HTML to DOCX conversions.
- [Health check](https://docs.case.dev/api-reference/superdoc/health-check.md): Unauthenticated health check for the SuperDoc service.
- [Populate DOCX template fields](https://docs.case.dev/api-reference/superdoc/populate-docx-template-fields.md): Populate fields inside a DOCX template using SuperDoc annotations. Supports text, images, dates, and numbers. Can target individual fields by ID or multiple fields by group.
- [Get API Status](https://docs.case.dev/api-reference/system/get-api-status.md): Returns basic information about the Case.dev API including name, version, and health status. This endpoint can be used for health checks and API discovery.
- [Health check](https://docs.case.dev/api-reference/system/health-check.md): Unauthenticated health check for the System service.
- [List documented services](https://docs.case.dev/api-reference/system/list-documented-services.md): Returns the public Case.dev services catalog derived from docs.case.dev/services. This endpoint is unauthenticated and intended for discovery surfaces such as the case.dev homepage.
- [Detect language](https://docs.case.dev/api-reference/translation/detect-language.md): Detect the language of text. Returns the most likely language code and confidence score. Supports batch detection for multiple texts.
- [List supported languages](https://docs.case.dev/api-reference/translation/list-supported-languages.md): Get the list of languages supported for translation. Optionally specify a target language to get translated language names.
- [Translate text](https://docs.case.dev/api-reference/translation/translate-text.md): Translate text between languages using Google Cloud Translation API. Supports 100+ languages, automatic language detection, HTML preservation, and batch translation.
- [Abort multipart upload](https://docs.case.dev/api-reference/vaults/abort-multipart-upload.md): Abort a multipart upload and discard uploaded parts (live).
- [Confirm vault upload result](https://docs.case.dev/api-reference/vaults/confirm-vault-upload-result.md): Confirm whether a direct-to-S3 vault upload succeeded or failed. This endpoint emits vault.upload.completed or vault.upload.failed events and is idempotent for repeated confirmations.
- [Create a new vault](https://docs.case.dev/api-reference/vaults/create-a-new-vault.md): Creates a new secure vault with dedicated S3 storage and vector search capabilities. Each vault provides isolated document storage with semantic search, OCR processing, and optional GraphRAG knowledge graph features for legal document analysis and discovery.
- [Create a vault share](https://docs.case.dev/api-reference/vaults/create-a-vault-share.md): Grant another organization read-only access to this vault. Only the vault owner can create shares. Idempotent: re-sharing to the same org updates the existing share.
- [Create vault group](https://docs.case.dev/api-reference/vaults/create-vault-group.md)
- [Create vault webhook subscription](https://docs.case.dev/api-reference/vaults/create-vault-webhook-subscription.md): Creates a webhook subscription for vault lifecycle events. Optional object filters can limit notifications to specific vault objects.
- [Delete vault](https://docs.case.dev/api-reference/vaults/delete-vault.md): Permanently deletes a vault and all its contents including documents, vectors, graph data, and S3 buckets. This operation cannot be undone. For large vaults, use the async=true query parameter to queue deletion in the background.
- [Delete vault group](https://docs.case.dev/api-reference/vaults/delete-vault-group.md)
- [Delete vault object](https://docs.case.dev/api-reference/vaults/delete-vault-object.md): Permanently deletes a document from the vault including all associated vectors, chunks, graph data, and the original file. This operation cannot be undone.
- [Delete vault webhook subscription](https://docs.case.dev/api-reference/vaults/delete-vault-webhook-subscription.md): Deactivates a vault webhook subscription.
- [Download vault object](https://docs.case.dev/api-reference/vaults/download-vault-object.md): Downloads a file from a vault. Returns the actual file content as a binary stream with appropriate headers for file download. Useful for retrieving contracts, depositions, case files, and other legal documents stored in your vault.
- [Generate presigned upload URL for vault file](https://docs.case.dev/api-reference/vaults/generate-presigned-upload-url-for-vault-file.md): Generate a presigned URL for uploading files directly to a vault's S3 storage. After uploading to S3, confirm the upload result via POST /vault/:vaultId/upload/:objectId/confirm before triggering ingestion.
- [Generate presigned URL for object operations](https://docs.case.dev/api-reference/vaults/generate-presigned-url-for-object-operations.md): Generate presigned URLs for direct S3 operations (GET, PUT, DELETE, HEAD) on vault objects. This allows secure, time-limited access to files without proxying through the API. Essential for large document uploads/downloads in legal workflows.
- [Get CaseMark job status](https://docs.case.dev/api-reference/vaults/get-casemark-job-status.md): Get the status of a CaseMark summary workflow job.
- [Get GraphRAG statistics](https://docs.case.dev/api-reference/vaults/get-graphrag-statistics.md): Retrieve GraphRAG (Graph Retrieval-Augmented Generation) statistics for a specific vault. This includes metrics about the knowledge graph structure, entity relationships, and processing status that enable advanced semantic search and AI-powered document analysis.
- [Get object text](https://docs.case.dev/api-reference/vaults/get-object-text.md): Retrieves the full extracted text content from a processed vault object. Returns the concatenated text from all chunks, useful for document review, analysis, or export. The object must have completed processing before text can be retrieved.
- [Get OCR word bounding boxes](https://docs.case.dev/api-reference/vaults/get-ocr-word-bounding-boxes.md): Retrieves word-level OCR bounding box data for a processed PDF document. Each word includes its text, normalized bounding box coordinates (0-1 range), confidence score, and global word index. Use this data to highlight specific text ranges in a PDF viewer based on word indices from search results.
- [Get presigned URLs for multipart upload parts](https://docs.case.dev/api-reference/vaults/get-presigned-urls-for-multipart-upload-parts.md): Generate presigned URLs for individual multipart upload parts (live).
- [Get vault information](https://docs.case.dev/api-reference/vaults/get-vault-information.md): Retrieve detailed information about a specific vault, including storage configuration, chunking strategy, and usage statistics. Returns vault metadata, bucket information, and vector storage details.
- [Get vault object with download URL](https://docs.case.dev/api-reference/vaults/get-vault-object-with-download-url.md): Retrieves metadata for a specific document in a vault and generates a temporary download URL. The download URL expires after 1 hour for security. This endpoint also updates the file size if it wasn't previously calculated.
- [Health check](https://docs.case.dev/api-reference/vaults/health-check.md): Unauthenticated health check for the Vault service.
- [Index object in GraphRAG](https://docs.case.dev/api-reference/vaults/index-object-in-graphrag.md): Manually trigger GraphRAG indexing for a vault object. The object must already be ingested (completed status). This extracts entities, relationships, and communities from the document for advanced knowledge graph queries.
- [Ingest vault object](https://docs.case.dev/api-reference/vaults/ingest-vault-object.md): Triggers ingestion workflow for a vault object to extract text, generate chunks, and create embeddings. For supported file types (PDF, DOCX, TXT, RTF, XML, ZIP, audio, video), processing happens asynchronously. ZIP archives are unpacked recursively up to 5 levels, and each extracted file is created as an independent vault object and ingested via the normal pipeline. For unsupported types (images, etc.), the file is marked as completed immediately without text extraction. GraphRAG indexing must be triggered separately via POST /vault/:id/graphrag/:objectId.
- [Initialize GraphRAG workspace](https://docs.case.dev/api-reference/vaults/initialize-graphrag-workspace.md): Initialize a GraphRAG workspace for a vault to enable advanced knowledge graph and retrieval-augmented generation capabilities. This creates the necessary infrastructure for semantic document analysis and graph-based querying within the vault.
- [List all vaults](https://docs.case.dev/api-reference/vaults/list-all-vaults.md): List all vaults for the authenticated organization. Returns vault metadata including name, description, storage configuration, and usage statistics.
- [List vault groups](https://docs.case.dev/api-reference/vaults/list-vault-groups.md)
- [List vault objects](https://docs.case.dev/api-reference/vaults/list-vault-objects.md): Retrieve all objects stored in a specific vault, including document metadata, ingestion status, and processing statistics.
- [List vault shares](https://docs.case.dev/api-reference/vaults/list-vault-shares.md): List all cross-org shares for a vault. Only the vault owner can list shares.
- [List vault webhook subscriptions](https://docs.case.dev/api-reference/vaults/list-vault-webhook-subscriptions.md): Lists webhook subscriptions configured for a vault.
- [Revoke a vault share](https://docs.case.dev/api-reference/vaults/revoke-a-vault-share.md): Revoke cross-org access to a vault. Only the vault owner can revoke shares.
- [Search vault documents](https://docs.case.dev/api-reference/vaults/search-vault-documents.md): Search across vault documents using multiple methods including hybrid vector + graph search, GraphRAG global search, entity-based search, and fast similarity search. Returns relevant documents and contextual answers based on the search method.
- [Send test webhook event](https://docs.case.dev/api-reference/vaults/send-test-webhook-event.md): Delivers a test event to a single vault webhook subscription. Uses the same payload shape, signature, and retry behavior as production event delivery.
- [Update vault group](https://docs.case.dev/api-reference/vaults/update-vault-group.md)
- [Update vault object](https://docs.case.dev/api-reference/vaults/update-vault-object.md): Update a document's filename, path, or metadata. Use this to rename files or organize them into virtual folders. The path is stored in metadata.path and can be used to build folder hierarchies in your application.
- [Update vault settings](https://docs.case.dev/api-reference/vaults/update-vault-settings.md): Update vault settings including name, description, and enableGraph. Changing enableGraph only affects future document uploads - existing documents retain their current graph/non-graph state.
- [Update vault webhook subscription](https://docs.case.dev/api-reference/vaults/update-vault-webhook-subscription.md): Updates callback URL, filters, active state, or signing secret for a vault webhook subscription.
- [Create transcription job](https://docs.case.dev/api-reference/voice/create-transcription-job.md): Creates an asynchronous transcription job for audio files. Supports two modes:

**Vault-based (recommended)**: Pass `vault_id` and `object_id` to transcribe audio from your vault. The transcript will automatically be saved back to the vault when complete.

**Direct URL (legacy)**: Pass `audio_url` for direct transcription without automatic storage.
- [Delete transcription](https://docs.case.dev/api-reference/voice/delete-transcription.md): Deletes a transcription job. For managed vault jobs (tr_*), also removes local job records and managed transcript result objects. Idempotent: returns success if already deleted.
- [Get streaming transcription WebSocket URL](https://docs.case.dev/api-reference/voice/get-streaming-transcription-websocket-url.md): Returns the WebSocket URL and connection details for real-time audio transcription. The returned URL can be used to establish a WebSocket connection for streaming audio data and receiving transcribed text in real-time.

**Audio Requirements:**
- Sample Rate: 16kHz
- Encoding: PCM 16-bit little-endian
- Channels: Mono (1 channel)

**Pricing:** $0.01 per minute ($0.60 per hour)
- [Get transcription status](https://docs.case.dev/api-reference/voice/get-transcription-status.md): Retrieve the status and result of an audio transcription job. For vault-based jobs, returns status and result_object_id when complete. For legacy direct URL jobs, returns the full transcription data.
- [Health check](https://docs.case.dev/api-reference/voice/health-check.md): Unauthenticated health check for the Voice service.
- [List available voices](https://docs.case.dev/api-reference/voice/list-available-voices.md): Retrieve a list of available voices for text-to-speech synthesis. This endpoint provides access to a comprehensive catalog of voices with various characteristics, languages, and styles suitable for legal document narration, client presentations, and accessibility purposes.
- [Stream text-to-speech audio](https://docs.case.dev/api-reference/voice/stream-text-to-speech-audio.md): Convert text to speech using ElevenLabs AI voices with streaming for real-time playback. Returns audio data as an MP3 stream for immediate playback with minimal latency. Perfect for legal document narration, client presentations, or accessibility features.
- [Text-to-speech synthesis](https://docs.case.dev/api-reference/voice/text-to-speech-synthesis.md): Convert text to natural-sounding audio using ElevenLabs voices. Ideal for creating audio summaries of legal documents, client presentations, or accessibility features. Supports multiple languages and voice customization.
- [Deploying Applications](https://docs.case.dev/compute/deploying.md): Put your app on the internet with one click—no servers, no configuration, no headaches
- [null](https://docs.case.dev/compute/functions.md)
- [Orbit Compute Overview](https://docs.case.dev/compute/index.md): GPU instances, dedicated AI models, and sovereign deployments
- [Projects API](https://docs.case.dev/compute/projects.md): Manage projects and deployments programmatically
- [Sovereign Legal AI](https://docs.case.dev/compute/sovereign-ai.md): Dedicated model deployments for clients with specific requirements
- [Contract Analysis](https://docs.case.dev/cookbooks/contract-analysis.md): Extract key terms, identify risks, and compare clauses across contracts
- [Deposition Workflow](https://docs.case.dev/cookbooks/deposition-workflow.md): Transcribe and analyze witness testimony with speaker labels, vault storage, and cross-examination reports
- [Discovery Pipeline](https://docs.case.dev/cookbooks/discovery-pipeline.md): Ingest documents, extract text via OCR, generate embeddings, and enable semantic search
- [Document Analyzer](https://docs.case.dev/cookbooks/document-analyzer.md): Search documents and generate AI-powered analysis with vault search and LLM
- [Document Intelligence Agent](https://docs.case.dev/cookbooks/document-intelligence-agent.md): Build a RAG-powered agent that can store and retrieve information from your documents
- [Cookbooks Overview](https://docs.case.dev/cookbooks/index.md): End-to-end tutorials and production patterns
- [Branches](https://docs.case.dev/databases/branches.md): Isolated database environments
- [Connections](https://docs.case.dev/databases/connection.md): Connection strings and pooling
- [Databases Overview](https://docs.case.dev/databases/index.md): Serverless PostgreSQL databases with branching
- [Projects](https://docs.case.dev/databases/projects.md): Create and manage database projects
- [Format](https://docs.case.dev/format/index.md): Generate professional PDF, DOCX, and HTML documents from structured content
- [Digitize Evidence](https://docs.case.dev/getting-started/digitize-evidence.md): Extract text from scanned documents
- [Discovery Search](https://docs.case.dev/getting-started/discovery-search.md): Search thousands of documents in seconds
- [Getting Started](https://docs.case.dev/getting-started/index.md): Build legal automation in 5 minutes
- [Deposition Prep](https://docs.case.dev/getting-started/litigation-depositions.md): Transcribe and analyze witness testimony
- [Web Research](https://docs.case.dev/getting-started/research-automation.md): AI-powered research with citations
- [Audit Logging](https://docs.case.dev/governance/audit.md): Track governance decisions and generate compliance reports
- [AI Governance Overview](https://docs.case.dev/governance/index.md): Control which AI providers your organization can use
- [Policy Configuration](https://docs.case.dev/governance/policies.md): Create and manage governance policies for your organization
- [Provider Trust Registry](https://docs.case.dev/governance/providers.md): Trust tiers and compliance certifications for all AI providers
- [The Toolkit for Legal Engineering](https://docs.case.dev/index.md): Enterprise primitives. Startup velocity.
- [Citations](https://docs.case.dev/legal-research/citations.md): Parse, verify, and retrieve legal citations
- [Docket Search](https://docs.case.dev/legal-research/docket-search.md): Search federal court dockets and filings via CourtListener RECAP
- [Legal Research Overview](https://docs.case.dev/legal-research/index.md): Search, analyze, and verify legal sources from authoritative databases
- [Patent Search](https://docs.case.dev/legal-research/patent-search.md): Search US patent applications and granted patents via the USPTO
- [Case Search](https://docs.case.dev/legal-research/search.md): Find cases by topic and discover related authorities
- [Trademark Search](https://docs.case.dev/legal-research/trademark-search.md): Look up US trademark status and details via the USPTO TSDR system
- [Chat completions](https://docs.case.dev/llms/chat-completions.md): Send messages to AI models and get responses
- [Embeddings](https://docs.case.dev/llms/embeddings.md): Convert text into numerical vectors for semantic search, similarity comparison, and clustering
- [LLMs Overview](https://docs.case.dev/llms/index.md): Access 195+ language models through one API
- [Model Catalog](https://docs.case.dev/llms/model-catalog.md): Browse all available AI models with pricing, capabilities, and context windows
- [List models](https://docs.case.dev/llms/models.md): Get all available AI models with pricing, context windows, and capabilities
- [Memory Overview](https://docs.case.dev/memory/index.md): Persistent memory for AI agents
- [Download results](https://docs.case.dev/ocr/download.md): Download OCR results as text, JSON, or searchable PDF
- [OCR Overview](https://docs.case.dev/ocr/index.md): Extract text from PDFs, images, and scanned documents
- [Process document](https://docs.case.dev/ocr/process.md): Extract text from PDFs, images, and scanned documents
- [Check status](https://docs.case.dev/ocr/status.md): Get OCR job status and progress
- [Payments](https://docs.case.dev/payments/index.md): Legal-grade payment infrastructure built for trust accounting, settlements, and compliance
- [Privilege Detection Overview](https://docs.case.dev/privilege/index.md): Detect attorney-client privilege and work product protection in documents for e-discovery
- [Rate Limits & Usage](https://docs.case.dev/rate-limits.md): Understanding Case.dev rate limits and usage tiers
- [SDKs](https://docs.case.dev/sdks.md): Official client libraries for the Case.dev API
- [All Services](https://docs.case.dev/services.md): Complete reference for all Case.dev platform services
- [Annotate Templates](https://docs.case.dev/superdoc/annotate.md): Populate fields in DOCX templates with dynamic data
- [Convert Documents](https://docs.case.dev/superdoc/convert.md): Convert documents between DOCX, PDF, Markdown, and HTML formats
- [SuperDoc Overview](https://docs.case.dev/superdoc/index.md): Convert documents between formats and populate DOCX templates with data
- [Thurgood](https://docs.case.dev/thurgood.md): Build legal apps by describing what you want—no coding required
- [Detect Language](https://docs.case.dev/translation/detect.md): Detect the language of text
- [Translation Overview](https://docs.case.dev/translation/index.md): Translate text between 100+ languages
- [Supported Languages](https://docs.case.dev/translation/languages.md): List all supported translation languages
- [Translate Text](https://docs.case.dev/translation/translate.md): Translate text between languages
- [Groups](https://docs.case.dev/vault/groups.md): Organize vaults into groups and scope API key access
- [Vaults Overview](https://docs.case.dev/vault/index.md): Document storage with AI-powered search
- [Manage](https://docs.case.dev/vault/manage.md): List vaults, download files, rename, move, and delete documents and vaults
- [Search](https://docs.case.dev/vault/search.md): Find documents by meaning—from simple queries to complex analysis
- [Upload & Process](https://docs.case.dev/vault/upload.md): Get documents into your vault and make them searchable
- [Webhooks](https://docs.case.dev/vault/webhooks.md): Receive vault ingestion lifecycle events without polling
- [Voice Overview](https://docs.case.dev/voice/index.md): Audio transcription and text-to-speech
- [Text-to-Speech](https://docs.case.dev/voice/speech.md): Generate natural-sounding speech from text
- [Transcription](https://docs.case.dev/voice/transcription.md): Convert audio and video to text
- [AI Answer](https://docs.case.dev/web-search/answer.md): Get AI-generated answers with citations
- [Web Search Overview](https://docs.case.dev/web-search/index.md): Search the web and generate AI answers with citations
- [Deep Research](https://docs.case.dev/web-search/research.md): Multi-step research for comprehensive analysis
- [Search](https://docs.case.dev/web-search/search.md): Search the web with domain filtering, date ranges, and content extraction

## OpenAPI Specs

- [openapi](https://docs.case.dev/openapi.json)

## Optional

- [Thurgood Code](https://thurgood.case.dev)