Available Tools
The AI Suite MCP currently provides 45 tools. Each tool belongs to exactly one OAuth scope, and a scope is only granted if the user's backend group has the appropriate AI Suite feature rights (see Permissions & Scopes). The tool list that a model sees is therefore different for each user. If a right is missing, the affected tool does not appear at all.
The following overview is grouped by scopes and briefly states the purpose of each tool. The complete parameters are provided by each client itself, as they are part of the tool definition in JSON schema. Which tools a specific installation actually delivers is revealed by readServerInfo.
Context and Research (mcp:read)
These tools only read. They do not cost any credits and do not call an AI model.
| Tool | Purpose |
|---|---|
readServerInfo | Status of the server with the versions of TYPO3, AI Suite, and MCP, the active configuration, and a diagnosis |
readPageTree | Traverse the page tree, limited to the user's mounts |
readPageContent | Read content elements of a page, optionally with container nesting |
readContentTree | Contents of all pages of a subtree at once, paginated by page |
readRenderedPage | The page as a visitor sees it, including plugin output. Requires additionally enable_mcp_rendered_page_read |
readEditorialGuidelines | The guidelines set by the editorial team regarding tone, target audience, and style for a page area |
readChildren | List container and IRRE children of a record, grouped by relation |
searchContent | Full-text search over pages and content elements |
listFiles | List files of a FAL storage or folder |
readFileInfo | Metadata of a single file |
listStaleContent | Find pages and content that have not been edited for N days |
readTaskStatus | Progress of a background task |
readTaskResults | Retrieve results of a completed task, read-only |
Records: Schema and Editing (mcp:read / mcp:write)
The exploratory tools of this group belong to mcp:read, the writing ones to mcp:write. All write operations go through the DataHandler and thus through the usual TYPO3 checks, and they follow the set write mode.
| Tool | Scope | Purpose |
|---|---|---|
listTables | mcp:read | List tables that the user is allowed to read, minus mcpExcludedTables |
readRecordSchema | mcp:read | TCA schema of a table with fields, types, validation, relations, and writeability |
readFlexFormSchema | mcp:read | Inner schema of a FlexForm field with sheets and fields |
listPageTypes | mcp:read | Available page types (Doktypes) |
listContentTypes | mcp:read | Available CTypes and valid columns of a page |
readRecords | mcp:read | Read records, by UID, by page, or by field filter |
compareWithLive | mcp:read | Field-wise comparison of a workspace draft with the live state |
previewRecords | mcp:write | Preview a write operation as old-new diff, without saving |
writeRecords | mcp:write | Create or modify records, optionally as an atomic batch |
copyRecords | mcp:write | Copy records, individually or as a batch |
moveRecords | mcp:write | Move records |
deleteRecords | mcp:write | Delete records (soft delete). Marked as destructive, the client will ask for confirmation |
localizeRecord | mcp:write | Create a translation wrapper for a record, without AI and without credits |
savePageTree | mcp:write | Save a generated page tree |
replaceText | mcp:write | A literal replacement in a field, without resending the entire field |
patchText | mcp:write | Multiple replacements in a field, applied atomically |
bulkReplaceText | mcp:write | The same replacement across all child records of a parent |
copyMediaReference | mcp:write | Copy file reference from one field to another |
replaceMediaReference | mcp:write | Replace the file behind an existing reference |
AI Functions (Generation, Translation, Images)
These tools call the providers and models of the AI Suite and consume credits. Which models are available is determined by the AI Suite configuration and the model rights of the backend group. No additional settings are required in the MCP for this.
They are therefore the only tools that require a valid API key from the AI Suite. If it is missing or the license package does not cover the function, the tools remain in the tool list and are also called by the model. The call then ends with a license error. This distinguishes them from the feature rights of the backend group, which can already remove a tool from the list.
| Tool | Scope | Purpose |
|---|---|---|
generateFileMetadata | mcp:generate | Generate alternative text, title, and description for a file, based on the file itself |
translateRecord | mcp:translate | Translate a single record |
translatePage | mcp:translate | Translate an entire page, metadata, and all content elements |
translateFileMetadata | mcp:translate | Translate file metadata into a target language |
generateImage | mcp:image | Generate an image from a text description and store it in FAL |
Plain language and the DeepL glossary of the site are not a separate tool, but part of the translation tools. The glossary is applied automatically.
Embed Media
This tool brings existing files into the file management without involving an AI model. It does not cost any credits and works even without a valid API key from the AI Suite. It is tied to its own scope and its own feature right, both turned off by default, because it is the only tool besides generateImage that creates a physical file and therefore cannot be reverted through a workspace.
| Tool | Scope | Purpose |
|---|---|---|
uploadMedia | mcp:media | Take existing images or videos into FAL, via URL, as Base64, or as a YouTube or Vimeo link |
Not to be confused with copyMediaReference and replaceMediaReference. The two do not insert a new file but transform existing file references, thus belonging to mcp:write and being part of the record group. For target folder, size limit, allowed file extensions, and securing remote downloads, see Security & Operation.
Background Tasks (mcp:workflow)
Batch tools run asynchronously. They immediately return a task ID, the progress is queried with readTaskStatus, and the results are read with readTaskResults. Nothing is written automatically. A batch generates suggestions, and only applyTaskResults writes them into the records.
| Tool | Purpose |
|---|---|
batchGenerateMetadata | Page metadata in large numbers, either for a UID list or for an entire page subtree |
batchGenerateFileMetadata | File metadata for a list of files |
batchGenerateFolderMetadata | File metadata for all files in a folder |
batchTranslatePage | Translate multiple pages |
batchTranslateFileMetadata | Translate file metadata of a list of files |
batchTranslateFolderMetadata | Translate file metadata of all files in a folder |
applyTaskResults | Write the translations of a completed batch into the localization records (Scope mcp:write) |
batchGenerateMetadata takes its targets in exactly one of two ways. Either pageIds as an explicit UID list or rootPageId as a page along with everything underneath. Both together is an error, and neither of them alone is. recursive decides whether a rootPageId traverses the entire subtree or stays with the direct children. The root page is always included.
Cost Limit for Batch Metadata
A rootPageId is limited to 50 pages. If there are more, the call is rejected before anything is billed, and the found page count is stated. The reason is an asymmetry. This tool costs credits per page, and a subtree is a quantity that no one has counted beforehand. "Everything below the starting page" is a short sentence away from a very large bill. An explicit pageIds list, on the other hand, is a quantity that has been consciously named and remains unlimited. Those who want to stay within the limit choose a deeper root page, set recursive to false, or pass the UIDs individually.