Instagram MCP Server

The Instagram MCP Server bridges the gap between AI assistants and live Instagram content by leveraging an existing Chrome login session. It enables developers to fetch, download, and process Instagram posts without the need for separate authentication flows or third‑party APIs. This is particularly valuable in contexts where AI models must reason about real‑time social media data, curate content for downstream tasks, or generate insights that require access to the actual media files and associated metadata.

At its core, the server exposes a single JSON‑RPC 2.0 compliant tool—. When invoked, the server uses a headless browser instance that reuses the user data directory provided by . This means the tool can access private profiles, direct messages, or any content the logged‑in user has permissions for. The tool accepts a username, an optional limit (1–50 or “all”), and optional parameters for media storage paths and pacing delays. Once the posts are retrieved, the server automatically downloads all media assets (images, videos, carousels) and generates a concise SEO‑friendly description for each post. The resulting dataset is returned as structured JSON, making it immediately consumable by downstream AI workflows.

Key capabilities include:

• Modular architecture: Clear separation of core MCP logic, Instagram‑specific features, and shared browser services ensures that each component can evolve independently.
• Type safety: Comprehensive TypeScript typings catch errors at compile time, improving developer confidence and IDE tooling.
• Robust error handling: Standardized JSON‑RPC error codes (, , etc.) provide predictable failure modes for AI assistants to handle gracefully.
• Configurable behavior: Environment variables and a central config manager allow fine‑tuning of browser viewport, timeouts, batch sizes, and output directories without code changes.
• Automatic media management: The server downloads files to a user‑specified directory and attaches metadata, enabling downstream tasks such as image analysis or content indexing.

Real‑world use cases span from automated social media monitoring—where an AI assistant can pull the latest posts of a brand for sentiment analysis—to content curation pipelines that ingest Instagram media into knowledge bases or generate AI‑powered summaries. By integrating directly with the MCP framework, developers can invoke this tool from any Claude or other AI assistant session, passing arguments as JSON and receiving structured results without leaving the conversational context. The server’s design prioritizes reliability, maintainability, and developer ergonomics, making it a standout choice for teams that need seamless, authenticated access to Instagram data within AI‑driven workflows.