Mem is a note-taking and knowledge management application that helps users capture, organize, and retrieve information efficiently.
Mem is a note-taking and knowledge management application that helps users capture, organize, and retrieve information efficiently. On Nagent, Mem is exposed as a fully-configurable notes integration that any agent can call — 11 actions, and API key authentication. No code is required to wire Mem into your workflow — connect it once via the External Integrations panel and reuse it across every agent you build.
Agent builders use Mem to automate the kinds of tasks notes teams previously handled manually. Concrete examples — each one is a single agent step in Nagent — include:
Every action and trigger is paired with a structured input/output schema (visible in the sections below), so when you wire Mem into Helix — our agentic agent builder — the editor knows exactly what each step expects and produces. Configure once, deploy anywhere across your Nagent agents.
Every operation an agent can call against Mem, with input parameters and output schema. Drop these into any step of an agent built in Helix.
MEM_CREATE_COLLECTIONCreates a new collection in Mem for organizing and grouping related notes. Collections are containers that help categorize notes by topic, project, or any organizational scheme. Each collection has a title and optional description. Use this action when you need to: - Create a new organizational container for notes - Set up a project workspace with a descriptive title - Organize notes by category or theme Returns the created collection's details including its unique ID.
Input parameters
Optional UUID for the collection. If not provided, a UUID will be auto-generated.
Title of the collection. Supports Markdown formatting (max ~1k characters).
Optional ISO 8601 datetime for the creation timestamp. Defaults to current time if not provided.
Optional ISO 8601 datetime for the last update timestamp. Defaults to created_at if not provided.
Optional description for the collection. Supports Markdown formatting (max ~10k characters).
Output
Data from the action execution
Error if any occurred during the execution of the action
Whether or not the action execution was successful or not
MEM_CREATE_NOTE_V2Tool to create a new note with markdown content and optional collection associations. The first line of content is automatically interpreted as the title. Use when you need to create a note and optionally add it to one or more collections by ID or title.
Input parameters
Custom note identifier (UUID); auto-generated if omitted
Markdown-formatted note body, max ~200k characters. The first line is automatically interpreted as the title.
ISO 8601 datetime for note creation timestamp; defaults to current time
ISO 8601 datetime for last modification timestamp; defaults to created_at value
Array of existing collection UUIDs to associate with the note
Array of collection names for association (case-insensitive, max ~1k chars each)
Output
Data from the action execution
Error if any occurred during the execution of the action
Whether or not the action execution was successful or not
MEM_DELETE_COLLECTIONTool to permanently delete a Mem collection. Deletion is irreversible — only invoke after explicit user confirmation and verification of the correct collection_id.
Input parameters
The UUID of the collection to delete. Must be a valid UUID format (e.g., '550e8400-e29b-41d4-a716-446655440000'). The deletion is idempotent - deleting a non-existent collection will still succeed.
Output
Data from the action execution
Error if any occurred during the execution of the action
Whether or not the action execution was successful or not
MEM_DELETE_NOTETool to permanently delete a specific note. Deletion is irreversible — obtain explicit user confirmation before calling. Use when you need to remove a note by its unique identifier after confirming the note_id.
Input parameters
UUID of the note to delete. Must be a valid UUID format (e.g., 'xxxxxxxx-xxxx-xxxx-xxxx-xxxxxxxxxxxx'). Obtain this from create_note or read_note responses.
Output
Data from the action execution
Error if any occurred during the execution of the action
Whether or not the action execution was successful or not
MEM_GET_COLLECTIONRetrieve the details of a Mem collection by its UUID. Returns the collection's title, description, and timestamps. Use this when you need to fetch metadata for a specific collection.
Input parameters
UUID of the collection to retrieve. Obtained from MEM_CREATE_COLLECTION or other collection listing actions.
Output
Data from the action execution
Error if any occurred during the execution of the action
Whether or not the action execution was successful or not
MEM_LIST_COLLECTIONSList collections with pagination support. Returns collections sorted by updated_at or created_at. Use this action to retrieve all collections or browse through collections page by page.
Input parameters
Opaque cursor token for pagination, returned from a previous request's next_page field. Omit this parameter for the first page of results.
Maximum number of collections to return per page. Defaults to 50 if not specified.
Sort order options for collection listing.
Output
Data from the action execution
Error if any occurred during the execution of the action
Whether or not the action execution was successful or not
MEM_LIST_NOTESTool to list notes with pagination and filtering options. Supports filtering by collection, task presence, image presence, and file presence. Use when you need to retrieve multiple notes or search for notes matching specific criteria.
Input parameters
Opaque cursor returned from a previous request. Omit for the first page.
Maximum number of notes to return. Defaults to 50.
Sort order for results. Choose between 'updated_at' (default) or 'created_at'.
Optional collection ID; when provided, only notes belonging to this collection are returned.
When true, only notes containing at least one file/attachment are returned.
When true, only notes containing at least one task (open or closed) are returned.
When true, only notes containing at least one image are returned.
When true, only notes containing at least one open task are returned.
When true, include the markdown content for each note in results.
Output
Data from the action execution
Error if any occurred during the execution of the action
Whether or not the action execution was successful or not
MEM_READ_NOTERetrieve the content and metadata of a Mem note by its UUID. Returns the note's title, markdown content, timestamps, and collection membership. Use this when you need to read or display an existing note's content.
Input parameters
UUID of the note to retrieve. Obtained from MEM_CREATE_NOTE or other note listing actions.
Output
Data from the action execution
Error if any occurred during the execution of the action
Whether or not the action execution was successful or not
MEM_SAVE_CONTENTTool to process and remember any raw content using AI. Accepts web pages, emails, transcripts, articles, or simple text. Use when you want to save and process content with optional instructions on how to process it and context about how it relates to existing knowledge.
Input parameters
Any raw content you want to remember - web pages, emails, transcripts, articles, or simple text. Common use cases: HTML from a webpage, email text (including headers), meeting transcript, text from a document, simple notes or thoughts. Maximum of ~1M characters (~1 MB).
Additional background information that helps Mem understand how this input relates to your existing knowledge. Examples: 'This is related to my Project Alpha research', 'This is from the quarterly planning meeting', 'This is a follow-up to our discussion yesterday'. Maximum of ~10k characters.
When this information was encountered or created (ISO 8601 datetime string). Defaults to the current date and time if not provided. Examples: '2025-04-01T14:30:45Z', '2023-12-15T09:45:30+01:00'.
Optional guidance on how you want this information processed or remembered. Examples: 'Extract the key findings and save as a research note', 'Create a summary with action items and decisions', 'Add to my reading list'. Maximum size: approximately 10 KB.
Output
Data from the action execution
Error if any occurred during the execution of the action
Whether or not the action execution was successful or not
MEM_SEARCH_COLLECTIONSTool to search collections using an optional query string. Use when you need to find or list collections by title or description.
Input parameters
Optional query string used to filter collections by title or description. If not provided, returns all collections.
Output
Data from the action execution
Error if any occurred during the execution of the action
Whether or not the action execution was successful or not
MEM_SEARCH_NOTESTool to search notes in Mem using a query string with optional filtering. Supports filtering by collection IDs, task presence, image presence, and file presence.
Input parameters
Maximum number of results to return per page (1-50, default 20).
Optional query string used to filter notes. When not provided, returns all notes matching the filters.
Configuration flags for the response payload.
Number of results to skip for pagination (default 0).
Snapshot ID from a previous search response, used to ensure consistent pagination across pages.
Optional list of collection IDs (UUIDs); when provided, only notes in any of the specified collections are returned.
When true, only notes containing at least one file/attachment are returned. Defaults to false.
When true, only notes containing at least one task (open or closed) are returned. Defaults to false.
When true, only notes containing at least one image are returned. Defaults to false.
When true, only notes containing at least one open task are returned. Defaults to false.
Output
Data from the action execution
Error if any occurred during the execution of the action
Whether or not the action execution was successful or not
No publicly available marketplace agent is found using this tool yet. There are 79 agents privately built on Nagent that already use Mem.
Build on Nagent
Connect Mem to any Nagent agent in minutes — no API key management, no boilerplate. Just configure and deploy.
The five questions agent builders ask before adopting a new integration.
Open the External Integrations panel inside Nagent (app.nagent.ai/externalIntegration), find Mem, and click "Connect Now." You'll authenticate with an API key — Nagent handles credential storage and refresh automatically. Once connected, Mem is available to any agent in your workspace.
No. Nagent provides no-code integration for every tool. Once Mem is connected, you configure its 11 actions directly in the agent builder UI — no API calls, no boilerplate, no schema management.
Helix — Nagent's agentic agent builder — lets you drop Mem steps into any workflow visually. Pick an action (e.g., one of those listed above), fill in the inputs (Helix knows the required vs. optional schema for each parameter), and connect it to upstream/downstream steps. Triggers run as the entry point of an agent, so when a Mem event fires, the agent kicks off automatically.
Every Mem action and trigger ships with a fully-typed schema — input parameters with name, type, required flag, and description, plus the output payload shape. The schemas are documented in the sections above. Helix uses these schemas to validate your configuration at build time and to type-check the data flowing between steps.
Yes. While Mem ships with 11 pre-built notes actions, you can layer custom logic around them inside Helix — pre/post-processing steps, conditional branches, retries, or stitching Mem together with other connected tools. For deeper customization, talk to our team about Nagent's Agentic AI Lab — forward-deployed engineers who build Mem-based workflows tailored to your business.