/
Use models, Agent, skills, research, retrieval, and private preferences.
Ask one plain-language question about a connected dataset and return an answer grounded in those rows.
Give the chat assistant inputs two ways: attach images/videos with a role (reference, first frame, last frame) or select existing canvas nodes — both get auto-wired into the node(s) the assistant crea
Analyze a connected research dataset for repeatable hook patterns instead of judging examples one by one.
Collect cited findings with a preset or custom brief and feed them into the rest of the Flowgen graph.
Run web research on a topic — Quick Search for a fast summary or Deep Research for a focused, multi-source report — and pipe the result as text/HTML into downstream generators.
Agentic nodes run a multi-step AI pipeline — reason about the best model (optionally researching), generate, then auto-QA the result — within a budget you set. Agentic Image and Agentic Video share th
Choose GLM 5.3 Flash for text, image, and video input; understand required reasoning, structured output, and the per-call price floor.
Switch the bottom workspace into Agent for conversational help, direct graph building, or an editable staged plan executed with the canonical Flowgen tools.
Choose the compact Make composer or the conversational Assistant, attach references with the right roles, review an editable staged plan, and inspect live node reasoning.
An LLM node that writes, sharpens or extracts text — pick Generate, Improve or Extract mode, choose an output format, and feed it any upstream media as typed context.
Build a bounded dataset of award-gallery references or direct URLs, including design signals and optional measured quality evidence.
Use Assist for conversation/direct work or Plan & Build for a reviewable multi-step graph, while keeping the host model and Flowgen tools responsible for execution.
Choose from the expanded model roster while capability-aware controls and live Flowie prices stay in sync.
Describe a multi-stage task and the assistant builds the whole pipeline at once — creating several nodes, wiring their outputs into later nodes, and running them in dependency order so frames generate
Build a bounded cross-platform dataset of posts for keywords or hashtags, with count, ordering, transcripts, and safe-search controls.
A chat panel that turns a plain description into wired, configured, and running nodes on the canvas. It opens from the assistant button inside the Make bar, and works in two modes: Assist, which adds
Choose between collecting social or website evidence, analyzing creative and performance patterns, or answering a question from an existing dataset.
flowGen routes each node to a wired provider: images (Freepik/Vertex), video (Veo, Luma, Runway, Kling), avatar (Hedra), voice (ElevenLabs/Google), music (Suno), 3D (Meshy).
Analyze engagement and content evidence from a connected dataset and emit structured performance findings plus a report.
Let Flowgen search Blueprints, Marketplace groups, shared projects, apps, and your own eligible flows for a proven graph shape, then insert and adapt it instead of rebuilding blindly.
Find profiles by niche in Discover mode or inspect named handles in Watch mode across selected platforms.
Automate your website’s image alt text and test AI prompt precision in one workflow. Learn how flowmo’s Photo Transcribe Flow extracts deep scene descriptions from images and verifies their accuracy t
Open a Video Editor node’s same-window compositor, discover its live native schemas, read and mutate the revisioned document, review evidence, and close it without starting Flowmo AI.
Use this overview to distinguish two Flowgen retrieval systems: discoverable reference graphs you can insert and adapt, and a creator-private preference profile used only as soft guidance.
Understand the private style profile inferred from your own Flowgen projects, how it influences suggestions, and why an explicit request always wins.
Use the bottom Make/Agent surface and the reorganized left rail without hunting through the old top-bar workflow.
Choose a built-in, imported, or owned reusable AI method, inspect what it writes, edit a safe owned copy, and wire its skill_out text into one or more compatible AI steps.
Use the bottom Make bar to choose a concrete action, attach compatible references, select a model, and add a ready-to-run node to the canvas in one transaction.
Choose reusable AI methods from the Prompt Library, place a Skill node when a method should be explicit in the graph, edit an owned copy safely, and understand how packages, skill sets, surface gates,
Scan selected research platforms for ranked trends around a topic or vertical, bounded by period and optional region.
Analyze a Website Inspiration dataset as two separate questions: what is visually worth learning from, and what is technically safe to copy.