AI Prompt

Runs one prompt through an AI model and returns the text reply — use it to generate, rewrite, answer, summarize, or analyze. Works with any OpenAI-compatible endpoint (OpenAI, OpenRouter, local Ollama at http://localhost:11434/v1, LM Studio) or Anthropic. The prompt comes from an input field or a {{ }} template; outputs the reply, finish reason, and token usage.

Finding it in the library

Search the builder's node library for AI Prompt (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.

Wired up in the builder

AI Prompt in a real, runnable flow — captured live from the Studio editor, exactly as it looks on your canvas. The ◈ ports carry the AI Model provider you wire in. This is the same workflow used for the example input & output below.

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.

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.

How data flows through it

AI Prompt 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 systemPrompt, prompt. 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 (AI Prompt) works as well as the exact type id (ai.prompt).

In the Copilot panel (or any connected AI):

add a ai prompt node after the trigger

As a step in a create_chain_workflow call:

{"type":"AI Prompt","config":{}}

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": "AI Prompt" } } }'

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

Output — what the node produced

Property reference

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

Using it

One completion, any OpenAI-compatible server

AI Prompt is the “one prompt in, one reply out” node. Point Base URL at any OpenAI-compatible endpoint — https://api.openai.com/v1, http://localhost:11434/v1 for a local Ollama, http://localhost:1234/v1 for LM Studio — set the Model, and pick where the prompt comes from:

• Prompt mode = field: read the prompt text from a dot-path on the input (Prompt field = q reads $json.q). Leave the field blank and the node sends the whole payload, stringified — handy for “summarize whatever arrives”.
• Prompt mode = template: build the prompt with {{ }} expressions, same semantics as the Set node — e.g. Summarize in one sentence: {{ $json.article }}.

The output (see the capture above) is a flat object: reply, the finish reason, and token usage — so the next node reads {{ $json.reply }} directly.

Auth without pasting keys

Pick a stored credential in the node’s credential picker (template openai-compat) — the API key field then carries ${credential.<name>.apiKey}, a vault reference resolved at run start, never the literal key. Local servers need no key at all: leave it blank.

Structured answers

Turn on JSON mode to request response_format: json_object — the node parses the reply into replyJson so downstream nodes can read fields without a script step. Pair it with a system prompt that names the fields you expect.

Tips

• Temperature is a string on purpose — leave it blank to use the server’s default rather than forcing one.
• Use the AI in Flowdrome guide (/docs/using/ai/) for the local-Ollama setup and a full worked example with screenshots.
• For multi-turn conversations, tool calling, or memory, step up to the AI Agent node — AI Prompt is deliberately stateless.

Related nodes

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

• AI Classify Classifies JSON content
• AI Agent Conversational AI agent with a pluggable model (Ollama local/cloud), pluggable memory (state/file/in-memory),
• AI Embeddings Turns text into embedding vectors using OpenAI-compatible endpoints or a local Ollama daemon
• AI Vector Store Self-contained semantic memory: embeds text internally and inserts, queries (cosine similarity), or clears an
• AI Extract Uses an LLM to pull structured fields (order number, customer, intent, dates, …) out of free text
• AI Eval Score an AI output against assertions: exact, contains, not-contains, regex, JSON-schema, and LLM-judge (a rub
• AI Transcribe Turn speech into text via any OpenAI-compatible /audio/transcriptions endpoint (Whisper, or a local faster-whi
• AI Speak Turn text into spoken audio (TTS) via any OpenAI-compatible /audio/speech endpoint

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. The hand-authored notes above are merged in.