Overview | Masset | Documentation

The Masset MCP Server provides access to content stored within Masset from external AI platforms such as Claude, ChatGPT, and other MCP-compatible clients. This allows for the decentralization of content use and consumption, while permitting Marketing and Sales teams to keep a pulse on what assets are being used. For more information about the Model Context Protocol, you can read the official MCP documentation.

What you can do

Using the Masset MCP Server, you can:

Search assets — find content by keyword, semantic meaning, or a combination of both. Apply filters by tags, collections, MIME types, and date ranges.
Retrieve metadata — get detailed file information including name, description, file type, size, properties, tags, colors, and detected objects.
Read asset content — access extracted text content including raw text, summaries, OCR text, and transcriptions.
Preview assets — view JPEG thumbnail previews inline in AI conversations.
Explore taxonomy — discover available tags, collections, and categories to refine your searches.
Analyze performance — rank assets by downloads, views, or shares over configurable time windows.
Track activity — get per-asset usage activity (views, downloads, shares), optionally scoped to specific users or groups.
Explore organizational context — list collections, boards, user groups, and users to resolve names into IDs for downstream tools.
Build and share content rooms — create, update, find, and list shareable content rooms backed by assets and/or boards, optionally tied to a HubSpot CRM contact, company, or deal.
Submit content requests — file a request for a new asset or a change to an existing asset; the request becomes a tracked task in the Masset app.
Manage training — discover, create, edit, and run training sessions; manage their sources, manual inputs, facts, and questions; AI-generate facts and questions; and read leaderboards and per-user stats.

Tools

The MCP server exposes 30 tools that AI clients auto-discover upon connection. Most are read-only; the content room, content request, and training mutation tools modify state in your Masset workspace.

Tool	Description	Read Only?
`asset_search`	Search for assets by keyword or meaning and return ranked results with metadata. Primary search interface.	Yes
`keyword_search`	Agentic use. Case-insensitive substring matching across asset names, descriptions, file names, tags, and more.	Yes
`semantic_search`	Agentic use. Meaning-based search via embeddings. Finds conceptually related content.	Yes
`asset_analytics_search`	Rank assets by an analytics metric (downloads, views, shares) over a time window.	Yes
`get_asset_metadata`	Retrieve detailed metadata for a list of assets by their IDs.	Yes
`get_asset_content`	Retrieve extracted text content of assets by their IDs.	Yes
`get_asset_preview`	Fetch a JPEG preview image for an asset.	Yes
`get_asset_activity`	Get per-asset usage activity (views, downloads, shares), optionally scoped to specific users.	Yes
`get_search_context`	Retrieve available search taxonomy (tags, collections, categories) for the current tenant.	Yes
`list_collections`	List collections visible to you, with their asset IDs.	Yes
`list_groups`	List user groups visible to you, with their member user IDs.	Yes
`list_users`	Look up user details by ID, or list all users in the company.	Yes
`list_boards`	List boards visible to you, with their names and descriptions.	Yes
`content_room_find`	Find content rooms by ID, CRM reference, or free-text CRM lookup.	Yes
`content_room_list`	List all content rooms owned by the current user.	Yes
`content_room_create`	Create a new shareable content room from assets and/or boards, optionally tied to a HubSpot CRM object.	No
`content_room_update`	Update an existing content room (assets, boards, CRM, expiration, name).	No
`content_request_create`	File a new-asset or asset-change request that becomes a task in the Masset app.	No
`training_session_list`	List training sessions in the company, optionally filtered by status.	Yes
`training_session_get`	Get full dashboard for a training session: stats, per-fact, and per-user progress.	Yes
`training_session_create`	Create a new training session in DRAFT status.	No
`training_session_update`	Update metadata, schedule, and audience of a DRAFT training session.	No
`training_session_lifecycle`	Transition a training session: start, pause, resume, complete, or cancel.	No
`training_sources_manage`	Attach or detach an asset, collection, or board as a source on a DRAFT session.	No
`training_manual_inputs_manage`	Create, update, or delete free-form manual inputs used for AI fact generation.	No
`training_facts_manage`	Create, update, or delete facts on a DRAFT session.	No
`training_questions_manage`	Create, update, or delete questions on a DRAFT session.	No
`training_generate`	AI-generate facts from a source, or questions from a session’s facts.	No
`training_leaderboard_get`	Get the company-wide or session-specific training leaderboard.	Yes
`training_user_stats_get`	Get per-user training stats across all sessions (streaks, points, rank).	Yes

The toolset will continue to grow to include more Masset functionality over time. If you have a question or would like to suggest new functionality, please contact support@getmasset.com.

Key details

Detail	Value
Protocol	Model Context Protocol (MCP)
Transport	Streamable HTTP
Endpoint	`https://app.getmasset.com/api/v1/mcp`
Authentication	OAuth 2.0 with JWT Bearer tokens
Rate limit	60 requests per minute per user
Status	General Availability

Next steps

Installation

Set up the Masset MCP Server in Claude, ChatGPT, or other clients.

Tools

Explore all available tools, their parameters, and usage details.

FAQs

Common questions about authentication, scopes, rate limits, and more.