anaksor-ocr
Sokheng · Kit
Reads Khmer and English text from documents, receipts and photos of signs, using a detector for text lines and a recogniser for each line. Suits developers who…
New
Webhook that returns an AI reading of a stock chart image
Newversion 1 · updated
Overview
An n8n workflow that waits for a stock or crypto chart image posted to a web address. An OpenAI model reads it for novice traders, and the answer goes back to the sender as plain text.
Webhook that returns an AI reading of a stock chart image is an n8n workflow in the Finance & accounting category, listed by Mindket. It is free. This is version 1, last updated on . Download it on this page or copy its JSON, then import it into an n8n canvas. Nothing runs until you activate it in n8n.
At a glance
- Runs in
- n8n Cloud: import the file into a new workflow there
- Your own n8n server: the same file and steps
- A browser extension or an app: post the image to the production address
- Anything else that can send a POST request: the same steps
Platform
How it works
Trigger
Web request
It starts when something posts a chart image to its web address. Set first:
- The OpenAI model in the node, if you want another
- A sender that posts the image to the production address
Actions
- The webhook accepts a POST request carrying the chart image and holds it open
- An OpenAI model reads the image and writes a technical analysis in plain words for novice traders
- The reply goes back to the sender as text, with a warning that it is not advice
Result
A plain-text reading of the chart, returned as the reply to the same request
Integrations
What you need before you start.
OpenAI
Language model
Reads the chart image and writes the analysis.
NeedsOpenAI API key
Workflow check
Checked by Mindket on 13 Nov 2026
- Steps
- 3 steps, starting when its webhook is called
- AI
- 1 step calls an AI model: OpenAI
- Sign-ins
- Signs in to: OpenAI
- API keys
- No API key is left in the file
- Step types
- Mindket’s checker reads all 3 of its step types
About
What it is
This n8n workflow reads a stock or crypto chart image and sends back a plain explanation of it. Your own tool posts the image to the workflow’s web address. An OpenAI model studies the chart like a technical analyst, explains in simple words which way the market may go and warns that this is not advice. The reply goes back to the sender as text. Three nodes run it, beside two sticky notes.
Who it suits
Developers and trading educators who want a chart reader behind their own button. A browser extension, a script or a phone shortcut can post the image, and the workflow does the rest. Skip it if you want ready buy or sell signals. The model gives a reading in words, and it can be wrong.
| Industry | Who uses it | How they use it | Example |
|---|---|---|---|
| Trading education | Course makers | Give students a plain reading of a chart they upload | A teacher posts a screenshot and shows the class what the model sees. |
| Finance content | Newsletter writers | Get a first reading to test against their own view | A writer posts a weekly chart and compares the reply with her notes. |
| Software | Developers of trading tools | Add a read this chart call to an extension or an app | A developer wires a screenshot button to the web address and shows the reply in a panel. |
| Retail investing | Hobby traders | Ask for a quick explanation of a chart they just saved | A trader sends a daily chart and reads the answer over coffee. |
You need a little n8n: one credential to add, and a sender that can post an image.
Where to use it
Any n8n that your sender can reach over the web will run it. The reply comes back in the same request.
- n8n Cloud: import the file into a new workflow there
- Your own n8n server: the same file and steps
- A browser extension or an app: post the image to the production address
- Anything else that can send a POST request: the same steps
Setup
Download saves a zip with workflow.json and Mindket’s setup notes in your downloads folder. Import the file, add one OpenAI credential and send a test image to the test address.
| Need | Detail |
|---|---|
| n8n | On n8n Cloud or on a server of your own |
| OpenAI | An API key |
| A sender | Anything that can post an image to a web address |
- In n8n, create a workflow, open its menu, choose Import from File and pick
workflow.json. - Add an OpenAI credential and select it in
OpenAI. - Open
OpenAIand check the model. It readsgpt-4o-mini. Pick the model you want to run. - Open
Webhookand copy its test address. - Post a chart image to that address, as the body of the request or as a form file named
data. - Read the answer in the response. When it works, save the workflow, turn it on and use the production address.
How to use it
Post a chart image to the webhook and read the reply. In test mode, start Webhook listening first, then send the image. The run reads the image, asks the model and returns its text.
| To change | Where |
|---|---|
| What the model says | The Text field of OpenAI |
| How long the answer runs | The maximum tokens option of OpenAI |
| The model | The model field of OpenAI |
| The form of the reply | Respond to Webhook |
| Who may call it | The authentication choice in Webhook |
Example
A trading teacher keeps a folder of chart screenshots for a class. She posts one to the webhook from her terminal and waits.
The reply comes back as plain text in simple words. It names the trend the model sees and where it expects the market to go, and it ends with a warning that this is not binding advice. She reads it to the class and asks who agrees. Check each reading against the chart yourself. A model can misread a scale or a label.
For newcomers and experts
If you are new. n8n draws a workflow as boxes joined by lines, and each box, a node, does one job. Here the first box waits for an image, the second asks the model and the third sends the answer back. A webhook is a web address that starts the workflow when something posts to it.
For experts. Webhook holds the request open until Respond to Webhook runs, so the sender gets the model’s text and not n8n’s default reply.
| Node | What it does | What to know |
|---|---|---|
Webhook | Accepts a POST request and waits for the response node | n8n stores an image sent as the request body in a binary field named data |
OpenAI | Runs the Analyze Image operation on that field | The field name data is the node’s default, and its maximum tokens fall back to 300 in n8n’s source |
Respond to Webhook | Sends $json.content back as text | Simplify output is on by default, so the node returns one content field |
Good to know
The webhook has no password and the model’s shutdown date is near, so look at both before you share the address.
- Model: the file sets
gpt-4o-mini. OpenAI’s deprecations page, read on 12 November 2026, gives 14 January 2027 as the shutdown ofgpt-4o,gpt-4o-mini,gpt-4.1andgpt-4.1-mini, withgpt-6-lunaas the replacement - No password: anyone who learns the address can post images and spend your OpenAI credit. Keep the address private, or turn on authentication in
Webhook - Short answers: the file sets no maximum tokens, so n8n’s default of 300 can cut a long reading short
- Not advice: the prompt asks the model to warn that its reading is not binding. Do not trade on it alone
- The sender: the workflow was built as the back end for a Chrome extension, and the extension is not in the file. You bring your own sender
- Repairs: the webhook path of the original copy was its author’s own, and Mindket’s copy holds a fresh random one
- Billing: OpenAI bills each image
- Your data: the chart image goes to OpenAI
- Credentials: the JSON holds none
README and files
Webhook that returns an AI reading of a stock chart image
An n8n workflow with three nodes. It listens on a webhook for an image of a stock or crypto price chart. An OpenAI
model reads it as a technical analyst would, and the answer goes back as the webhook's reply. Anything that can post
an image to a web address can use it: a browser extension, a script or a phone shortcut.
What is inside
workflow.json: the workflow as n8n exports it, 3 nodes and 2 sticky notes. It holds no credential.README.md: these notes.
What we changed from the original copy
- The webhook's path was the author's own. It now holds a fresh random one. Your own address appears once you import the
file, and you can change the path inWebhook. - The two sticky notes held a long description and a link to the author's pages. They now hold one short title each, and
the second points to the prompt inOpenAI. - Nothing else changed. The model and the prompt are as they were. See Good to know for the model's shutdown date.
Before you start
- An n8n instance, on n8n Cloud or your own server.
- An OpenAI API key.
- Something that posts a chart image to the webhook. The workflow holds no sender.
Set it up
- In n8n, create a workflow, open its menu and choose Import from File. Pick
workflow.json. - Add an OpenAI credential and select it in
OpenAI. - Open
OpenAIand look at the model. It readsgpt-4o-mini. Pick the model you want to run. - Open
Webhook. It shows a test address and a production address. Use the test address while you build. - Send a chart image to the test address with a POST request, as the body of the request or as a form file named
data, then read the answer in the response. For example, with curl:curl -X POST --data-binary @chart.png -H "Content-Type: image/png" <YOUR_TEST_URL> - When it works, save the workflow, turn it on and use the production address.
How a run goes
Webhookaccepts a POST request. It answers later, throughRespond to Webhook, so the sender waits for the reply.
n8n stores an image sent as the request body in a binary field calleddata.OpenAIruns the Analyze Image operation on that binary field. The text it sends along is the prompt: it asks for a
technical analysis of the chart for novice traders, in plain words, with a warning that this is not advice.Respond to Webhooksends back the model's reply as plain text.
Change the look
- What the model says: the Text field of
OpenAIholds the whole prompt. - How long the answer is: the node's options hold a maximum tokens setting. It is not set in the file, and n8n's source
falls back to 300 tokens, so a long reading can end early. Add the option and raise it. - The model:
OpenAI, the model field. - The reply's form:
Respond to Webhooksends plain text. Change it to JSON to return the answer with other fields.
Good to know
- The facts below come from pages and n8n source read on 12 November 2026.
- The model is
gpt-4o-mini. OpenAI's deprecations page gives 14 January 2027 as the shutdown date ofgpt-4o,gpt-4o-mini,gpt-4.1andgpt-4.1-mini, and namesgpt-6-lunaas their replacement. Pick another model in the node
before that date. - The webhook has no authentication. Anyone who learns its address can send images and spend your OpenAI credit. Keep the
address private, or put a check in front of the node. - The prompt asks the model to warn that its reading is not binding advice, and a model's reading of a chart is a guess.
Do not trade on it alone. - The workflow was built as the back end for a Chrome extension that posts a chart. The extension is not in this file, so
you bring your own sender. - Billing: OpenAI bills each image, and the picture's size changes the cost.
- Your data: the chart image goes to OpenAI.
- Credentials: the JSON holds none.
NOTICE.md 654 B README.md 4 KB workflow.json 3 KB
What’s new
Reviews
More from Mindket
See all from MindketQwen2.5-7B-Instruct-GGUF by Qwen
Qwen · Kit
Qwen2.5-7B-Instruct as GGUF files in eight quantization sizes, to chat with locally through llama.cpp. For people who want a capable 7.6B model on their own co…
New
ACE-Step 1.5 XL SFT
ACE-Step · Kit
Generates music from text with a 4B diffusion decoder, 50 steps and guidance control, aiming for higher audio quality than the base. For creators with a GPU of…
New
Similar in Finance & accounting
See all in Finance & accountingKronos-mini
NeoQuasar · Kit
Forecasts financial candlestick series with a 4.1M-parameter foundation model trained on data from over 45 exchanges. For quant researchers and developers who…
New
An n8n workflow that sends a Hebrew WhatsApp reminder for an unpaid invoice 7, 14 and 30 days after its due date. It reads the open invoices from a Google Shee…
NewFree
An n8n workflow that files invoices. It watches a Google Drive folder, has LlamaParse read each new file and an OpenAI model pick out the line items. Airtable…
NewFree