MCP Tools Reference

Strata exposes 17 tools via the Model Context Protocol, organized into groups. By default, only core tools are enabled. Enable additional groups via the X-Strata-Tool-Groups header.

Tool Groups

Coredefault

Document read, edit, list, and search

app_get_section_contentedit_documentfindget_document_graphget_imageread_document

Collaboration

Section locking and presence

get_presence

Agents

Platform agent invocation

get_agent_statusinvoke_agent

Header Configuration

Add the X-Strata-Tool-Groups header to your MCP client configuration to enable tool groups. Use comma-separated group names, or "all" to enable everything.

json

{
  "url": "https://api.strata.space/mcp",
  "headers": {
    "X-Strata-Tool-Groups": "core,spaces,agents"
  }
}

Core

default

Document read, edit, list, and search

#`app_get_section_content`

Get the full content of a document section by document ID and section ID. Returns the section text and heading context.

Parameter	Type	Description
`chunkId`	`string`	The chunk ID to retrieve
`documentId`	`string`	The document ID containing the section

#`edit_document`

Edit a document, manage access, or upload content from a file.

ACTIONS - "create": title, [content], [folderId] — create with inline content. - "edit": documentId, oldString, newString, [replaceAll] — targeted replacement. - "write": documentId, newContent — full INLINE replacement. - "share": documentId, email, role — grant access. - "unshare": documentId, targetUserId — revoke access. - "presignUpload": contentLength + (documentId OR title [+folderId]). Returns { uploadId, uploadUrl, requiredHeaders, expiresAt }. - "finalizeUpload": uploadId. Idempotent. Returns { documentId, bytes, created }.

Upload flow (prefer when content already exists as a file or as bytes you generated — avoids re-emitting them as tool input): 1. wc -c file.md → byte count 2. presignUpload(contentLength, ...) → { uploadUrl } 3. PUT file to uploadUrl with requiredHeaders 4. finalizeUpload(uploadId) → { documentId }

Parameter	Type	Description
`action`	`string`	Action to perform (see tool description)
`content`?	`string`	Initial markdown body (action `create`)
`contentLength`?	`integer (uint64)`	Byte count of payload (action `presignUpload`)
`documentId`?	`string`	Document ID
`email`?	`string`	Recipient email (action `share`)
`folderId`?	`string`	Folder ID (actions `create`, `presignUpload`)
`newContent`?	`string`	Full new body (action `write`)
`newString`?	`string`	Replacement text (action `edit`)
`oldString`?	`string`	Text to find (action `edit`)
`replaceAll`?	`boolean`	Replace every match, default false (action `edit`)
`role`?	`string`	viewer, commenter, or editor (action `share`)
`targetUserId`?	`string`	User to revoke (action `unshare`)
`title`?	`string`	Document title (actions `create`, `presignUpload`)
`uploadId`?	`string`	ID from `presignUpload` (action `finalizeUpload`)

#`find`

Search, list, or discover documents and spaces across the workspace. Use this tool whenever you need to locate content before reading it — set intent to the kind of discovery (search, list, recent, related, info) and scope to where to look (all, a specific space, folder, or document, the set of all spaces, published documents, or strataDocumentation for Strata's own product docs). Search uses hybrid keyword + semantic retrieval fused via Reciprocal Rank Fusion, so both short keywords and natural sentences work as queries. Use filters for status, owner, date ranges, includeShared, direction, sinceDays (for intent='recent'), and tokenBudget. Returns a uniform shape with items that carry id, title, snippet, score, and source attribution; follow up with read_document to fetch full content for any item you want to use. If you receive zero results, broaden the scope before reformulating the query — narrow scopes can return empty even when content exists.

Parameter	Type	Description
`intent`	`unknown`	Discovery intent. "search" ranks by relevance, "list" enumerates, "recent" returns newest changes within filters.sinceDays, "related" finds documents similar to scope.id (requires scope.type="document"), "info" returns metadata about the scoped entity.
`scope`	`unknown`	Where to look. Use IDs from the prompt or a prior find result.
`filters`?	`object`	Narrowing filters. Ignored when inapplicable to the chosen intent. For intent="recent", set filters.sinceDays (default 7).
`limit`?	`integer (uint32)`	Max results. Default: 10. Max: 50.
`nextToken`?	`string`	Pagination cursor from a prior find result. Pass through unchanged.
`query`?	`string`	Search text. Required when intent="search".
`tokenBudget`?	`integer (uint32)`	Per-chunk content size budget for intent="search". Default: server-chosen.

#`get_document_graph`

Get the connection graph neighborhood around a document. Returns nodes (documents) and edges (connections) discovered by BFS traversal.

Optional parameters: - depth: BFS depth (default 2, max 5) - maxNodes: Maximum nodes to return (default 50, max 200)

Parameter	Type	Description
`documentId`	`string`	The document ID to start the graph traversal from
`depth`?	`integer (uint32)`	BFS traversal depth (default 2, max 5)
`maxNodes`?	`integer (uint32)`	Maximum number of nodes to return (default 50, max 200)

#`get_image`

Retrieve an image from a document as base64-encoded data. Use when you need to analyze, describe, or understand an image referenced in a document.

The imageId is found in document content as [Image: alt text (imageId: <id>)] markers or ![alt](strata://image/<id>) in markdown.

Returns the image as inline base64 PNG data that can be viewed directly.

Parameter	Type	Description
`documentId`	`string`	Document the image belongs to. The caller must have `DocumentRead` on this document before bytes are returned
`imageId`	`string`	Image ID (SHA256 hex from document content)
`maxWidth`?	`integer (uint32)`	Maximum width to resize to before encoding (saves tokens). Defaults to 1024

#`read_document`

Read a document's content by ID.

Returns rendered Markdown. Use before editing. Large documents automatically return a structural outline instead of full content (default budget: 10000 tokens). Use the sections parameter with heading paths from the outline to read specific sections.

Workflow: read_document(id) → outline (if large) → read_document(id, sections: ["Heading > Subheading"]) → section content.

Parameter	Type	Description
`documentId`	`string`	Document ID to read, like `doc_01JAB...`
`commentFormat`?	`string`	Comment output format: "footnotes" (default, inline markdown footnotes) or "structured" (appends a JSON block with full comment thread data)
`includeComments`?	`boolean`	Whether to include comments in the response. Comments are rendered as inline footnotes in the markdown output
`includeSuggestions`?	`boolean`	Whether to include pending suggestions as a JSON block appended to the markdown content
`maxTokens`?	`integer (uint32)`	Token budget (default: 10000). Documents exceeding this return an outline instead of full content. Set to 0 to always get the outline
`sections`?	`string[]`	Heading paths to read, like `["Architecture > Database Layer"]`. Prefix matching: `"Architecture"` returns that section and all sub-sections. Use paths from a prior outline response

Collaboration

Section locking and presence

#`get_presence`

Query who is currently viewing or editing a document. Returns a list of collaborators with their entity type, display name, active section, status, and presence color. Useful for understanding who else is working on the document before making edits.

Parameter	Type	Description
`documentId`	`string`	Document ID to query presence for

Agents

Platform agent invocation

#`get_agent_status`

Check the status of an agent task.

Returns the current status, tool call count, and output (if completed) or error (if failed). Poll this after invoke_agent to track completion.

Possible statuses: pending, running, completed, failed, cancelled, timedOut.

Parameter	Type	Description
`taskId`	`string`	The task ID returned by `invoke_agent`

#`invoke_agent`

Invoke a platform agent to perform a task.

Specify the agent by ID or by name (mutually exclusive). Optionally provide a target document and a natural-language instruction.

Returns a task ID — use get_agent_status to poll for status and output. The agent executes asynchronously; this tool returns immediately after the task is created.

Parameter	Type	Description
`agentId`?	`string`	Agent ID to invoke (mutually exclusive with `agent_name`)
`agentName`?	`string`	Agent name to invoke (mutually exclusive with `agent_id`)
`documentId`?	`string`	Target document ID (optional)
`idempotencyKey`?	`string`	Idempotency key (optional)
`instruction`?	`string`	Natural-language instruction for the agent

MCP Tools Reference

Tool Groups

Header Configuration

Core

#app_get_section_content

#edit_document

#find

#get_document_graph

#get_image

#read_document

Collaboration

#get_presence

Agents

#get_agent_status

#invoke_agent

#`app_get_section_content`

#`edit_document`

#`find`

#`get_document_graph`

#`get_image`

#`read_document`

#`get_presence`

#`get_agent_status`

#`invoke_agent`