VideoVector technical documentation

Start here for the platform-level model: indexes, media, extraction engines, extraction executions, search, connectors, exports, and webhook-driven workflows.

• Platform model
• Extraction schema design
• Extraction execution model
• Search model

Use the guides section for task-oriented implementation steps that build on the core concepts and link directly into API, SDK, and MCP reference material.

• Create and rotate API keys
• Create extraction engines with nested schemas
• Add video-level synthesis
• Run extraction executions on an index, selected media, or playground content

The API reference is organized by resource type and workflow boundary for media ingestion, extraction, retrieval, automation, delivery, and agentic search.

• Auth and API keys
• Indexes and media
• Prompts
• Prompt runs

The Python SDK gives application teams typed access to VideoVector indexes, media, prompts, prompt runs, search, connectors, imports, exports, webhooks, and playground workflows.

• SDK setup for media workflows
• Workflow examples

The VideoVector MCP docs show how AI clients can browse indexes, run extraction engines, search media evidence, inspect workflow resources, and validate available tools.

• Claude Desktop setup
• Cursor setup
• Custom clients and HTTP transport
• Tools, helper endpoints, and playground