@remotion/mcp

Exposes Remotion's video composition framework as an MCP server that AI agents can discover and interact with via standardized protocol. Implements MCP server lifecycle management (initialization, resource listing, tool registration) to bridge Remotion's React-based composition API with LLM tool-calling systems, enabling agents to understand available composition patterns, rendering options, and media handling capabilities without direct SDK knowledge.

Scans Remotion's composition registry and exposes available templates, component patterns, and composition metadata as MCP resources with structured schemas. Implements resource enumeration that maps Remotion's internal composition structure (timeline duration, frame rate, dimensions, media dependencies) into queryable MCP resources, allowing agents to understand what compositions exist and their constraints before attempting to render or modify them.

Exposes Remotion rendering options (codec, bitrate, frame rate, resolution, output format) as MCP tools with JSON Schema validation. Implements tool schema generation that maps Remotion's RenderMediaOnLambda and local rendering APIs into callable MCP tools, with built-in parameter validation ensuring agents can only invoke valid rendering configurations and preventing malformed render requests that would fail downstream.

Provides MCP tools for resolving and validating media asset paths (video, audio, images) that Remotion compositions consume and produce. Implements path normalization, file existence checking, and format validation against Remotion's supported media types (H.264, WebM, PNG, JPEG, etc.), allowing agents to verify asset availability and compatibility before passing them to composition rendering without manual file system inspection.

Exposes Remotion's AWS Lambda and Google Cloud Run rendering backends as MCP tools with job submission, status tracking, and result retrieval. Implements tool wrappers around RenderMediaOnLambda and cloud-specific APIs that handle authentication, job queuing, and asynchronous result polling, allowing agents to submit long-running render jobs and check completion status without blocking or requiring direct cloud SDK knowledge.

Analyzes Remotion composition React component signatures and generates JSON Schema representations of their props, exposing these schemas as MCP resources. Implements TypeScript/JSDoc parsing to extract prop types, default values, and constraints, then converts them to JSON Schema for agent consumption, enabling LLMs to understand what parameters each composition accepts without reading source code or maintaining separate documentation.

Provides MCP tools for querying and manipulating Remotion's timeline system (frame numbers, duration, frame rate, sequence composition). Implements helpers that convert between human-readable time formats (seconds, milliseconds) and frame numbers, and expose Remotion's Sequence and Timeline APIs as callable tools, enabling agents to understand and construct complex multi-clip compositions without manual frame calculation.

Exposes Remotion's supported audio and video codecs (H.264, VP8, VP9, AAC, MP3, etc.) as MCP resources with quality presets and bitrate recommendations. Implements codec compatibility checking and preset generation based on target platform (web, mobile, social media) and quality requirements, allowing agents to select appropriate codecs without understanding compression trade-offs or platform-specific constraints.

Implements MCP tools that analyze rendering failures and provide diagnostic information and retry strategies. Parses Remotion rendering error messages and FFmpeg logs to identify root causes (missing assets, codec incompatibility, insufficient memory, timeout), then suggests corrective actions (retry with different codec, increase timeout, check asset paths), enabling agents to autonomously recover from transient failures or provide meaningful error context to users.

Provides MCP tools that profile Remotion composition rendering performance (frame render time, memory usage, CPU utilization) and suggest optimizations. Implements performance metric collection during test renders and analysis that identifies bottlenecks (expensive animations, large media assets, complex effects), then recommends optimizations (reduce animation complexity, compress assets, use memoization), enabling agents to optimize compositions for faster rendering without manual performance testing.