Agent Memory

ai.memory AI v0.1.0

Conversation memory as a node: choose the storage (state = durable host store, file = a JSONL per session you can open, inmemory = ephemeral), the session key, and how many turns to recall — then wire this node into the ◈ Memory port of any AI Agent. The storage settings flow into every agent it is attached to.

The Agent Memory step on the Studio canvas — The **Agent Memory** step as it appears on the Studio canvas — input pins on the left, output ports on the right.

Finding it in the library

Search the builder's node library for Agent Memory (it lives under AI). A single click opens the in-editor docs panel shown here — description, ports, and every property, without leaving the canvas. Double-click (or drag) to add it to the workflow.

Agent Memory in the node library, with the in-editor docs panel open — The library entry and the in-editor docs panel for **Agent Memory** — the same reference this page is generated from.

Wired up in the builder

Agent Memory in a real, runnable flow — captured live from the Studio editor, exactly as it looks on your canvas. This is the same workflow used for the example input & output below.

Agent Memory wired into a runnable workflow in the Studio builder — **Agent Memory** wired into a runnable flow — input on the left, output on the right.

How it’s configured

The node’s Configure panel as it opens in the builder when you select the step — every setting laid out with real values. Click any field to edit it.

The Agent Memory node's Configure panel in the Studio builder — The **Configure** panel for **Agent Memory**, showing the settings from the flow above.

Ports

Ports are the node’s contract with its neighbours. In the editor a port label renders bold when wired and italic when optional; ◈ ports accept attachment carriers rather than data wires.

Direction	Port	Label	What flows through it
Output	`output`	Memory

How data flows through it

Agent Memory consumes the content of the incoming envelope — when it is fed directly by a trigger, the trigger’s wrapper is unwrapped at the node boundary so the node sees the actual data, not the metadata shell. Its output becomes the payload for the next node, while the envelope (trace ids, correlation, binary refs) rides along untouched. In the Runs view you always see the whole envelope for both sides of this node.

Expressions in the config

String-typed properties accept {{ }} expressions evaluated against the incoming item at run time — e.g. {{ $json.customer.email }}. On this node that’s memoryDir, memoryEmbedModel, memoryEmbedBaseUrl, memoryEmbedKey, sessionKey. JSON- and code-typed fields never interpolate — they are passed through literally.

Build it with AI

Every node in this reference is reachable through Flowdrome’s AI Copilot and the MCP tools — say what you want, and the graph surgery happens server-side. Node types resolve fuzzily, so the catalog label (Agent Memory) works as well as the exact type id (ai.memory).

In the Copilot panel (or any connected AI):

add a agent memory node after the trigger

As a step in a create_chain_workflow call:

{"type":"Agent Memory","config":{}}

Raw MCP call — add this node to a workflow with add_node

curl -s -X POST http://localhost:48170/mcp -H "content-type: application/json" -d '{ "jsonrpc": "2.0", "id": "1", "method": "tools/call", "params": { "name": "add_node", "arguments": { "workflowId": "<id>", "type": "Agent Memory" } } }'

Example input & output

Captured from a real test run of the workflow above — this is what you see in the run data panel after pressing Test workflow.

Input — what the node received

The Agent Memory node's input envelope in the run data viewer — The input envelope in the Runs view — Flowdrome always shows the whole envelope, with the payload inside `body`.

Output — what the node produced

The Agent Memory node's output envelope in the run data viewer — The output envelope after the step ran.

Property reference

Every setting, with its type and default — the same fields shown configured in the panel above.

Property	Type	Default	Description
Storage `memoryBackend`	`select`	`"file"`	Where the conversation lives: state = the durable host store (survives restarts, invisible); file = a JSONL file per session (memoryDir); inmemory = ephemeral; vector = SEMANTIC recall (embeds each turn, recalls the most RELEVANT past exchanges instead of the last N). `fileinmemorystatevector`
Folder `memoryDir`	`string`	`"./agent-memory"`	Where the per-session .jsonl transcripts are written (file storage). Shown when `(memoryBackend ?? "file") === "file"`
Embed provider `memoryEmbedProvider`	`select`	`"ollama"`	Embeddings provider for vector memory: ollama (local) or openai (any OpenAI-compatible /embeddings). `ollamaopenai` Shown when `memoryBackend === "vector"`
Embed model `memoryEmbedModel`	`string`	`"nomic-embed-text"`	Embedding model for vector memory, e.g. nomic-embed-text (ollama) or text-embedding-3-small (openai). Shown when `memoryBackend === "vector"`
Embed URL `memoryEmbedBaseUrl`	`string`	`""`	Override the embeddings endpoint. Blank = the provider default. Shown when `memoryBackend === "vector"`
Embed key `memoryEmbedKey`	`string`	`""`	Bearer key for the embeddings endpoint (openai) — supports ${credential.NAME.FIELD}. Shown when `memoryBackend === "vector"`
Session key `sessionKey`	`string`	`"{{ input.sessionId }}"`	Template for the conversation id, e.g. {{ input.sessionId }}. Same key = same remembered conversation.
Recall turns `historyLimit`	`int`	`20`	How many prior turns the agent recalls each time.

Related nodes

The rest of the AI group — the same folder you’d scan in the editor’s library.

This page is generated from the node registry by gen-node-docs.mjs on every site build — ports, properties, defaults and visibility rules cannot drift from the code. The screenshots and example data are captured from a live Flowdrome by npm run shots:nodes and npm run gen:examples.