OpenRouter + Facebook, on-brand posts published fast

You sit down to “just post something,” and suddenly you’re juggling captions, translations, hashtags, and an image that doesn’t match your page style. Then you rewrite it twice because the tone feels off. Again.

This OpenRouter Facebook posts automation hits social media managers hardest, but course creators and small team owners feel it too. You will turn a single vocab word into a ready-to-publish lesson post (text plus visuals) without starting from a blank page every time.

Below, you’ll see how the workflow runs in n8n, what it outputs, and what you need to publish consistently on Facebook without babysitting the process.

How This Automation Works

The full n8n workflow, from trigger to final output:

n8n Workflow Template: OpenRouter + Facebook, on-brand posts published fast

Click to explore

flowchart LR

    subgraph sg0["When chat message received Flow"]
        direction LR
        n0["<div style='background:#f5f5f5;padding:10px;border-radius:8px;display:inline-block;border:1px solid #e0e0e0'><img src='https://flowpast.com/wp-content/uploads/n8n-workflow-icons/httprequest.dark.svg' width='40' height='40' /></div><br/>Generate Image (Recraft.ai)"]
        n1@{ icon: "mdi:robot", form: "rounded", label: "Create FB Post Content", pos: "b", h: 48 }
        n2@{ icon: "mdi:robot", form: "rounded", label: "Structured Output Parser2", pos: "b", h: 48 }
        n3@{ icon: "mdi:robot", form: "rounded", label: "Create Image Prompt", pos: "b", h: 48 }
        n5@{ icon: "mdi:brain", form: "rounded", label: "OpenRouter Chat Model", pos: "b", h: 48 }
        n6@{ icon: "mdi:brain", form: "rounded", label: "OpenRouter Chat Model1", pos: "b", h: 48 }
        n7["<div style='background:#f5f5f5;padding:10px;border-radius:8px;display:inline-block;border:1px solid #e0e0e0'><img src='https://flowpast.com/wp-content/uploads/n8n-workflow-icons/facebook.svg' width='40' height='40' /></div><br/>Facebook Graph API"]
        n8@{ icon: "mdi:play-circle", form: "rounded", label: "When chat message received", pos: "b", h: 48 }
        n9["<div style='background:#f5f5f5;padding:10px;border-radius:8px;display:inline-block;border:1px solid #e0e0e0'><img src='https://flowpast.com/wp-content/uploads/n8n-workflow-icons/code.svg' width='40' height='40' /></div><br/>Code"]
        n9 --> n1
        n3 --> n0
        n5 -.-> n1
        n1 --> n3
        n6 -.-> n3
        n2 -.-> n3
        n8 --> n9
        n0 --> n7
    end

    subgraph sg1["Flow 2"]
        direction LR
        n4["<div style='background:#f5f5f5;padding:10px;border-radius:8px;display:inline-block;border:1px solid #e0e0e0'><img src='https://flowpast.com/wp-content/uploads/n8n-workflow-icons/httprequest.dark.svg' width='40' height='40' /></div><br/>Text Over Image"]
    end

    %% Styling
    classDef trigger fill:#e8f5e9,stroke:#388e3c,stroke-width:2px
    classDef ai fill:#e3f2fd,stroke:#1976d2,stroke-width:2px
    classDef aiModel fill:#e8eaf6,stroke:#3f51b5,stroke-width:2px
    classDef decision fill:#fff8e1,stroke:#f9a825,stroke-width:2px
    classDef database fill:#fce4ec,stroke:#c2185b,stroke-width:2px
    classDef api fill:#fff3e0,stroke:#e65100,stroke-width:2px
    classDef code fill:#f3e5f5,stroke:#7b1fa2,stroke-width:2px
    classDef disabled stroke-dasharray: 5 5,opacity: 0.5
    class n8 trigger
    class n1,n2,n3 ai
    class n5,n6 aiModel
    class n0,n7,n4 api
    class n9 code
    class n4 disabled
    classDef customIcon fill:none,stroke:none
    class n0,n7,n9,n4 customIcon

The Problem: Facebook posts take forever to make consistent

Educational posts look simple from the outside, but the “simple” part is what drains you. One vocab word becomes a hook, a structured mini-lesson, examples, a pro-tip, a call-to-action, and hashtags. Then you need a matching visual that doesn’t feel like random clipart. If you’re teaching Chinese words to a Thai-speaking audience, you also have to keep the formatting tight (Chinese characters, Pinyin, Thai meaning) or the post becomes confusing fast. And frankly, consistency is what’s hardest to maintain when you’re rushing.

The friction compounds. Here’s where it breaks down most weeks.

It’s easy to burn about 45 minutes on one “quick” lesson post once you include writing, formatting, and image hunting.
The voice changes from post to post, so the page feels less professional even when the content is good.
Visuals don’t match your theme, which means you end up redesigning or skipping images altogether.
Publishing is another manual step, so posts slip when you’re busy with real work.

The Solution: Generate a lesson post + image, then publish automatically

This n8n workflow turns a single input word into a polished Facebook post that follows a repeatable structure and stays on-brand. It starts when you submit a vocab word through a chat trigger (you can also swap in Google Sheets later). OpenRouter generates the Thai-language lesson caption with the pieces you actually need: hook, the Chinese word with Pinyin, Thai meaning, usage examples, a quick pro-tip, and a call-to-action plus hashtags. Then a second OpenRouter prompt creates a clean image concept that fits the lesson. Recraft.ai generates the artwork in a consistent illustration style, and the workflow optionally applies a text overlay so your visuals look like a series, not a one-off. Finally, n8n publishes the caption and image to your Facebook Page using the Graph API.

The workflow begins with a word you provide. AI writes the caption and the image brief, Recraft renders the visual, and Facebook Graph API posts it to your page. One input, one finished post, live on Facebook.

What You Get: Automation vs. Results

What This Workflow Automates

Results You’ll Get

It turns one vocab input into a complete Thai lesson caption with a consistent structure.
It generates an image concept automatically so visuals match the lesson theme.
It creates a consistent illustration-style image via Recraft.ai (no design tool needed).
It publishes the final post to your Facebook Page through the Graph API.

Most teams get about 3 hours back each week if they post daily.
Your posts stop feeling “random” because the format stays consistent across the series.
You reduce last-minute rewriting since the hook, CTA, and hashtags are baked in.
Visuals look like a brand system instead of whatever you found that day.
Publishing becomes reliable, even when your calendar is chaotic.

Example: What This Looks Like

Say you publish 5 vocab posts per week. Manually, you might spend about 30 minutes writing and formatting the Thai lesson, plus another 15 minutes finding or making an image, plus 5 minutes posting and checking it, which is roughly 50 minutes per post (around 4 hours a week). With this workflow, you drop in the word in under a minute, wait about 2 minutes for AI text plus image generation, and the post is published automatically. You’re mainly reviewing, not building from scratch, so you get back about 3 hours most weeks.

What You’ll Need

n8n instance (try n8n Cloud free)
Self-hosting option if you prefer (Hostinger works well)
OpenRouter for generating the caption and image concept.
Recraft.ai to generate consistent lesson artwork.
Facebook Graph API access (get it from Meta for Developers and your Page settings).

Skill level: Intermediate. You’ll connect APIs, paste keys, and confirm Facebook Page permissions in Meta.

Don’t want to set this up yourself? Talk to an automation expert (free 15-minute consultation).

How It Works

You send a word to the workflow. The automation starts from an Incoming Chat Trigger, then a parsing step saves the vocab word in a clean format so it doesn’t get lost later.

AI drafts the lesson caption. OpenRouter generates a structured Thai post that includes the hook, the Chinese word plus Pinyin, Thai meaning, examples, and engagement prompts. This is where you lock in your “series” format, so every post feels familiar.

AI describes the visual, then an image gets rendered. A second OpenRouter prompt creates a short image concept, then an HTTP request calls Recraft.ai to generate the artwork in your predefined style. If you want the post to look even more consistent, the workflow can apply a simple overlay to the image too.

The final post gets published to Facebook. n8n sends the caption and the generated image to Facebook using the Pages API, so it lands on your Page without manual uploading.

You can easily modify the input source to use Google Sheets instead of chat based on your needs. See the full implementation guide below for customization options.

Step-by-Step Implementation Guide

Step 1: Configure the Chat Trigger

This workflow starts when a chat message is received, so you’ll configure the trigger that captures incoming chat input.

Add and open Incoming Chat Trigger.
Keep the default settings in Incoming Chat Trigger (no parameters are required).
Connect Incoming Chat Trigger to Parse Chat Input to pass the chat payload downstream.

Use the chat input test panel to send a sample message and confirm it appears in the trigger payload.

Step 2: Parse the Incoming Message

The workflow uses a code node to standardize the input for downstream AI and rendering steps.

Open Parse Chat Input and set JavaScript Code to the provided snippet that maps $json.chatInput into word and input.
Ensure the output JSON includes word and input as shown in the node’s code.
Connect Parse Chat Input to Draft Social Post.

Step 3: Set Up the Social Post Generation

Draft Social Post uses a chat model to generate the Facebook post content based on the parsed input.

Open Draft Social Post and set Text to =the word: {{ $json.word }}.
Keep Prompt Type as define and leave Has Output Parser enabled.
Open Primary Chat Model and set Model to google/gemini-2.5-pro.
Credential Required: Connect your openRouterApi credentials in Primary Chat Model.
Ensure Primary Chat Model is connected as the language model for Draft Social Post.

The long system message in Draft Social Post defines tone and structure—edit it only if you want a different voice or format.

Step 4: Build the Image Prompt with Structured Output

This step transforms the social post into a short image prompt and parses it into structured JSON.

Open Compose Image Prompt and set Text to ={{ $('Parse Chat Input').item.json.input }}.
Keep Prompt Type as define and leave Has Output Parser enabled.
Open Parse Prompt JSON and keep the JSON Schema Example as { "image_prompt_brief": "YOUR_GENERATED_SENTENCE_HERE" }.
Open Secondary Chat Model and set Model to google/gemini-flash-1.5.
Credential Required: Connect your openRouterApi credentials in Secondary Chat Model.
Ensure Secondary Chat Model is connected as the language model for Compose Image Prompt, and Parse Prompt JSON is connected as its output parser.

Credential Required: Output parsers like Parse Prompt JSON do not take credentials directly—add credentials to the parent node (Compose Image Prompt) via its connected model.

Step 5: Render the Artwork and Publish to Facebook

The prompt is sent to an image API, and the resulting image URL plus the post copy are published to Facebook.

Open Render Artwork and set URL to https://external.api.recraft.ai/v1/images/generations.
Set Method to POST and keep Specify Body as json.
Set JSON Body to { "prompt": "{{ $json.output.image_prompt_brief }}", "style": "digital_illustration", "substyle": "young_adult_book_2", "size": "1024x1024" }.
Credential Required: Connect your httpHeaderAuth credentials in Render Artwork.
Open Publish to Facebook and set Edge to photos and Node to [YOUR_ID].
Set Host URL to =graph.facebook.com and Graph API Version to v22.0.
In Query Parameters, set message to ={{ $('Draft Social Post').item.json.output }} and url to ={{ $json.data[0].url }}.
Credential Required: Connect your facebookGraphApi credentials in Publish to Facebook.

⚠️ Common Pitfall: Replace [YOUR_ID] with your Facebook Page ID, or the post will fail.

Step 6: Review Optional Utilities and Notes

The workflow includes a disabled overlay utility and a branding note for documentation.

Review Flowpast Branding as a documentation-only node; it does not affect execution.
If you plan to enable Utility: Image Text Overlay, configure the body parameters and image/text expressions as provided.
Credential Required: If you enable Utility: Image Text Overlay, connect your httpHeaderAuth credentials.

The overlay node is currently disabled, so it does not run in the execution flow.

Step 7: Test and Activate Your Workflow

Validate the workflow end-to-end before turning it on for production use.

Click Execute Workflow and send a sample chat message through Incoming Chat Trigger.
Confirm Draft Social Post outputs formatted content and Compose Image Prompt returns image_prompt_brief.
Verify Render Artwork returns an image URL in $json.data[0].url.
Check Facebook to ensure Publish to Facebook posts the image with the generated message.
When everything works, toggle the workflow Active to enable production execution.

🔒

Unlock Full Step-by-Step Guide

Get the complete implementation guide + downloadable template

Common Gotchas

Facebook Graph API credentials can expire or need specific permissions. If things break, check your Page token validity and permissions in Meta for Developers first.
If you’re using Wait nodes or external rendering, processing times vary. Bump up the wait duration if downstream nodes fail on empty responses.
Default prompts in AI nodes are generic. Add your brand voice early or you’ll be editing outputs forever.

Frequently Asked Questions

How long does it take to set up this OpenRouter Facebook posts automation?

About 30 minutes if you already have your API keys and Facebook Page access.

Do I need coding skills to automate OpenRouter Facebook posts?

No. You’ll connect accounts, add API keys, and paste your Facebook Page details into the right fields.

Is n8n free to use for this OpenRouter Facebook posts workflow?

Yes. n8n has a free self-hosted option and a free trial on n8n Cloud. Cloud plans start at $20/month for higher volume. You’ll also need to factor in OpenRouter and Recraft.ai API costs.

Where can I host n8n to run this automation?

Two options: n8n Cloud (managed, easiest setup) or self-hosting on a VPS. For self-hosting, Hostinger VPS is affordable and handles n8n well. Self-hosting gives you unlimited executions but requires basic server management.

Can I customize this OpenRouter Facebook posts automation for Google Sheets input instead of chat?

Yes, and it’s a common tweak. Replace the Incoming Chat Trigger with a Google Sheets trigger (new row or scheduled fetch), then map the “word” field into the same place the workflow currently uses after “Parse Chat Input.” Most people also add a simple “status” column in Sheets (Queued, Posted, Error) so nothing gets double-published. If you want approvals, route drafts to Slack first, then publish only when approved.

Why is my Facebook Graph API connection failing in this workflow?

Usually it’s a token or permission issue. Make sure your app and Page token include permissions like pages_manage_posts and the Page you’re targeting is the one tied to that token. Also confirm the page_id and Graph API version match what your Facebook node is sending. If it worked once and stopped, your token may have expired and needs to be regenerated.

How many posts can this OpenRouter Facebook posts automation handle?

A lot, as long as your API limits and n8n plan match your volume.

Is this OpenRouter Facebook posts automation better than using Zapier or Make?

For this kind of workflow, n8n is usually the better fit because it handles multi-step AI generation, branching, and HTTP calls without getting awkward or expensive as the logic grows. You also get the option to self-host, which matters when you’re posting frequently. Zapier or Make can be fine for simpler “one prompt, one post” setups, but you’ll feel the limits once you add structured parsing, image generation, and retries. If you’re unsure, Talk to an automation expert and describe your posting volume and review process.

Once this is running, you’re no longer “making a post.” You’re feeding a system that publishes clean, consistent lessons on command.

OpenRouter + Facebook, on-brand posts published fast

How This Automation Works

n8n Workflow Template: OpenRouter + Facebook, on-brand posts published fast

The Problem: Facebook posts take forever to make consistent