---
title: "Placid Integration"
description: "Generate on-brand creatives from Placid templates. Agents fill named template layers with dynamic text and images to render social images, multi-page PDFs, and videos, polling asynchronous renders until the output URL is ready."
url: https://flowrunner.ai/integrations/placid-app
date_modified: 2026-08-04T10:04:16-07:00
---

# Placid

[Developer Tools](https://flowrunner.ai/integrations/category/developer-infrastructure)

Generate on-brand creatives from Placid templates. Agents fill named template layers with dynamic text and images to render social images, multi-page PDFs, and videos, polling asynchronous renders until the output URL is ready.

[Verified](https://flowrunner.ai/integrations/verified "What does verified mean?") · 7 actions · API key · available

[Placid website](https://placid.app/) · [Platform Documentation](https://placid.app/docs) · Capability data verified 2026-08-12

1.  A new article publishes on the \[Webflow\](/integrations/webflow) site
2.  List Templates resolves the social banner template's uuid and its layer names
3.  Generate Image fills the title, author, and hero layers, with Create Now returning the finished URL in one call
4.  The agent confirms the render finished and that no layer fell back to empty or placeholder content
5.  The content lead approves the banner, the one check a template cannot make: whether this headline works at this crop
6.  The approved image attaches to the post's social metadata
7.  The marketing channel in \[Slack\](/integrations/slack) gets the finished banner and its post link

## What This Integration Enables

Placid draws a line most creative automation ignores: designers own the template, and everything else only fills its named layers. That split is what makes brand consistency programmable. A FlowRunner agent cannot nudge a logo or break the grid, because the design is locked in the template; it can only supply text and images to the layers the designer exposed. On that foundation, agents render social images, multi-page PDFs, and videos on demand, polling asynchronous renders until the output URL is ready. What automation cannot judge, whether the data in those layers should reach a customer, stays [human-in-the-loop](https://flowrunner.ai/concepts/human-in-the-loop).

-   Render on-brand social images and Open Graph banners from post and product data
-   Produce branded PDFs, from one-page certificates to multi-page documents, as data payloads
-   Compose short videos from video templates, with clips sequenced from multiple segments
-   Discover templates and their layer names programmatically, so flows adapt as designers iterate
-   Correlate every render to its source record through passthrough metadata and success webhooks

### Without FlowRunner

**Every asset is a design request**: Routine banners and cards queue behind real design work and ship late or never

**Brand drift, one remake at a time**: Each team rebuilds the template slightly wrong, and the feed slowly stops matching

**Documents assembled by hand**: Certificates and branded PDFs mean copy-paste sessions in a doc editor

### With FlowRunner

**Templates render themselves**: Named layers fill from live data the moment the triggering event happens

**The designer keeps control**: Design changes happen in the Placid template once and propagate to every render after

**Documents are a data payload**: Multi-page PDFs render from an array of pages, each one a template plus its data

## Use Case Scenarios

### Social banners that keep up with publishing

Every article that goes live on [Webflow](https://flowrunner.ai/integrations/webflow) gets its Open Graph banner rendered from the house template: title layer, author layer, hero image layer. Generate Image with Create Now returns the finished URL synchronously, the content lead approves in [Slack](https://flowrunner.ai/integrations/slack), and the banner attaches to the post before the first share. When the design team refreshes the template next quarter, every future banner ships in the new style with zero workflow changes, because the flow only ever knew the layer names, and the layer names did not move.

### Branded order documents at volume

Orders pull from [Shopify](https://flowrunner.ai/integrations/shopify) and each becomes a branded PDF: the agent fills the invoice template's layers with the order's line items, queues Generate PDF, and polls Get PDF until each document's URL is ready. Rendered documents log against their orders in [Google Sheets](https://flowrunner.ai/integrations/google-sheets), and the batch waits for a person before anything customer-facing sends, because these are financial documents with the company's name on them. The template guarantees they look right; the review guarantees they are right. Success webhooks carry each document's passthrough metadata back, so the flow always knows which order a finished PDF belongs to without a lookup.

### Certificates and a launch video from the same pipeline

A cohort finishes the certification course and the completion list arrives from [Typeform](https://flowrunner.ai/integrations/typeform). The agent renders a personalized certificate per graduate with Generate Image, then assembles the cohort announcement with Generate Video, sequencing template segments through the Clips array with each graduate's name card. Get Video polls until the render finishes, the program lead reviews the cut, and certificates go out by [Gmail](https://flowrunner.ai/integrations/gmail-service) while the video posts to the community. Two very different deliverables, one pipeline, and the only manual step left is the one that deserves a person: checking that names are spelled right on documents people will frame.

## Human-in-Loop Highlight

The template model means a Placid render can never be off-brand, which concentrates all the risk in one place: the data. Generate PDF filling an invoice template with the wrong amount produces a flawless, professional, wrong financial document, and once it reaches a customer's inbox it cannot be recalled, only corrected in a follow-up nobody enjoys sending. So customer-facing document runs are gated on the payload, not the pixels: the agent presents the rendered PDFs alongside the source records they were filled from, and a person confirms the amounts, names, and dates match before the batch sends. The same gate holds for Generate Video announcements, where a wrong date in a launch video is public the moment it posts. The agent's job is to make the review fast: rendered output and source data, side by side, one decision per document.

Agent processes routinely

Detects exception requiring judgment

Clear match Continues automatically

Ambiguous Routes to human via preferred channel

Human decides

Agent resumes with decision

## Agent Capabilities

7 actions

### Images

2

-   **Generate Image** Renders an image from a template by filling its named layers with text and image content, with layers also hideable per render. Create Now returns the finished image URL synchronously; otherwise the job queues for polling. Supports success webhooks, passthrough metadata, and output modifications like filename, size, and format.
-   **Get Image** Polls a queued image render by id until the status reaches finished and the image URL is populated, or reports the error.

### PDFs

2

-   **Generate PDF** Renders a PDF from templates: a Pages array of template-plus-layers objects for multi-page documents, or a single template and layers for one page. Queued for polling, with webhook and metadata options.
-   **Get PDF** Polls a queued PDF render by id, returning the downloadable PDF URL once finished.

### Videos

2

-   **Generate Video** Renders a video from a video template's named layers, with an optional Clips array to sequence multiple template segments into one cut. Queued for polling.
-   **Get Video** Polls a queued video render by id until the video URL is ready.

### Templates

1

-   **List Templates** Lists the project's templates with uuid, title, tags, and thumbnail. The lookup that gives flows a template's uuid and the layer names its renders must fill.

## Frequently Asked Questions

### What can FlowRunner do with Placid?

FlowRunner agents can run Generate Image, Get Image, and Generate PDF in Placid, plus 4 more actions.

### Does connecting Placid to FlowRunner require OAuth?

No. Placid connects to FlowRunner with an API key, no OAuth flow required.

### Can Placid trigger a FlowRunner workflow automatically?

Placid doesn't currently expose triggers in FlowRunner. It connects as an action step inside workflows started by another trigger.

**Work at Placid?** This integration exposes Placid to AI agents on every FlowRunner plan, including through MCP, at no cost to you. [See what FlowRunner offers integration partners](https://flowrunner.ai/integrations/partners), including how to keep this page current.

---
Markdown version of https://flowrunner.ai/integrations/placid-app. Site index: https://flowrunner.ai/llms.txt
