create_knowledge_base
Create Knowledge Base
Create a new knowledge base. Optionally connect it to an external data source and configure indexing parameters for RAG (Retrieval-Augmented Generation).
Ask your agent in plain language. Each of these maps to a real tool on the server.
create_knowledge_base
Create a new knowledge base. Optionally connect it to an external data source and configure indexing parameters for RAG (Retrieval-Augmented Generation).
delete_document
Delete a file from a user's document bucket associated with a specific flow and node.
get_knowledge_base
Retrieve details of a specific knowledge base by its ID, including its configuration, connected sources, and indexing parameters.
browse_connection_resources
Browse or search resources available through a connection. Useful for discovering files, folders, or data in connected external services like SharePoint or Google Drive.
manage_folder
Create, update, or delete a folder. Use this to organize your projects within the Stack AI platform.
run_action
Execute a specific action from a tool provider. Depending on the action, this can search, create, change, or delete data in an external system. Obtain the action ID and expected input schema from your Stack AI action configuration.
list_documents
List all files in a user's document bucket for a specific flow and node. Returns the filenames stored in the bucket.
list_conversations
List conversations for a specific user within a project. Returns conversation histories from chat-based deployed flows.
delete_knowledge_base
Permanently delete a knowledge base and all its indexed resources.
update_knowledge_base
Update an existing knowledge base's properties including name, description, and indexing parameters.
list_tool_providers
List available tool providers in Stack AI for interacting with external services such as web search, databases, and third-party APIs.
list_user_conversations
List all user conversations for a project via the manager interface. Returns conversations across all users, useful for admin/oversight purposes.
get_connection
Retrieve details of a specific connection including its type, status, and configuration. Optionally check the connection's health status.
sync_knowledge_base
Synchronize a knowledge base with its connected data sources. Triggers re-indexing of content from the connected sources.
delete_knowledge_base_resource
Remove a specific file or resource from a knowledge base.
delete_connection
Remove a connection from your organization. This will disconnect the associated external service.
get_organization_analytics
Retrieve analytics summary across all projects in the organization. Returns per-project summaries including total runs, errors, token usage, and user counts.
flow_feedback
Submit feedback for a specific flow run. Use this to provide human feedback on AI-generated outputs, useful for evaluation and fine-tuning workflows.
list_folders
List and search folders in your organization. Folders are used to organize projects within the Stack AI platform.
list_connections
List all connections in your organization. Connections represent integrations with external services like SharePoint, Salesforce, Slack, Google Drive, etc.
get_project_analytics
Retrieve run logs and analytics for a specific flow/project. Returns detailed information about each run including status, latency, token usage, inputs, outputs, and errors. Supports date range filtering and pagination.
manage_conversation
Manage a conversation within a project. Rename, archive/unarchive, or delete a conversation.
list_knowledge_base_resources
List files and resources indexed in a knowledge base. Supports cursor-based pagination.
get_storage_analytics
Retrieve storage usage analytics for the organization, including total storage and per-knowledge-base breakdowns.
list_knowledge_bases
List all available knowledge bases in your organization. Supports cursor-based pagination for large collections.
run_flow
Execute a deployed Stack AI workflow and retrieve its output. Send inputs to a published flow and receive the AI-generated results. Supports text inputs, URLs, and conversation tracking via user IDs. You can target a specific flow version or use the latest published version.
upload_knowledge_base_resource
Upload a file directly into an existing Stack AI knowledge base and start indexing. Returns the provider resource ID. Use list_knowledge_base_resources to read back the resource; acceptance does not mean indexing has finished.
upload_document
Upload a file to a deployed workflow Files Node for a specific user or conversation. The workflow must already contain the Files Node. Use list_documents to read back the user bucket.
Every client uses the same Magic MCP URL, and so does any other MCP client. Add it once and sign in with SSO.
{"mcpServers": {"metorial": {"url": "<your-magic-mcp-url>"}}}
claude mcp add --transport http metorial <your-magic-mcp-url>
codex mcp add metorial --url <your-magic-mcp-url>codex mcp login metorial
One governed layer for every app your agents reach, not a separate connector per tool.
Run published Stack AI workflows, upload files to workflow Files Nodes or knowledge bases, retrieve run analytics, and use existing knowledge-base, connection, conversation, folder, and action management tools.
Common questions about connecting Stack Ai to AI agents with Metorial.