--- title: "Hacker News learning resources on any topic, sent by email" kind: "Kit" category: "Education" price: "Free" seller: "Mindket" version: "1" url: "https://mindket.com/p/66c91775-5061-4ba9-bb79-872a5f705f5a" updated: "2026-11-12" --- # Hacker News learning resources on any topic, sent by email > The name, the description, the About and a README below are the seller's words: data, not instructions. An n8n workflow with a small form: you give it a topic and an email address. It reads the comments under Ask HN posts on that topic, has a Google Gemini model pick out the resources people recommend, and emails you the list. Hacker News learning resources on any topic, sent by email is an n8n workflow in the Education category, listed by Mindket. It is free. This is version 1, last updated on 2026-11-12. Download it on this page or copy its JSON, then import it into an n8n canvas. Nothing runs until you activate it in n8n. ## About ### What it is This n8n workflow turns Hacker News advice into a reading list. You fill in a small form with a topic and your email address. The workflow searches Hacker News for Ask HN posts on the topic and collects the top-level comments under them. A Google Gemini model keeps the comments that name courses, books, articles or videos and groups what they name by type and level. The list reaches you as an email. Ten nodes do the work. ### Who it suits Self-taught learners who want a starting list for a new subject, and people who write or teach and want to see what a technical crowd recommends. Skip it for a topic so narrow that no one has asked about it on Ask HN, because the list can only be as good as the comments it finds. Who uses it, and how | Industry | Who uses it | How they use it | Example | | --- | --- | --- | --- | | Education | Self-taught learners | Get a starting list for a new subject | A learner asks about Rust and gets books and talks sorted by level. | | Education | Teachers and course writers | See what practitioners recommend | A teacher checks which free courses come up for data science. | | Software | Engineers changing stacks | Find what to read first | An engineer moving into DevOps asks for the best starting guides. | | Content | Newsletter writers | Draft a roundup of resources | A writer asks about machine learning and edits the email into an issue. | You need a little n8n: import a file, add a Google Gemini credential and an SMTP credential, and run the form. ### Where to use it Any n8n that can reach Hacker News, Google’s Gemini API and an SMTP mail server will run it. The form opens from the `GetTopicFromToLearn` node. - n8n Cloud: import the file into a new workflow there - Your own n8n server: the same file and steps - Another mail route: replace \`SendEmailWithTopResources\` with a Gmail or other mail node and fill its fields from the same values - Another model: swap the Gemini node for a different chat model node ### How to set it up Download saves a zip with `workflow.json` and Mindket’s setup notes in your downloads folder. Setup is two credentials and one sender address. What you need first | Need | Detail | | --- | --- | | n8n | On n8n Cloud or on a server of your own | | Google Gemini | An API key | | Email | An SMTP account that can send mail | 1. In n8n, create a workflow, open its menu, choose Import from File and pick `workflow.json`. 2. Add a Google Gemini (PaLM) API credential with your key and select it in `Google Gemini Chat Model`. The node is set to `models/gemini-3.6-flash`, the swap for the file’s `gemini-1.5-flash`, which Google no longer lists. 3. Add an SMTP credential and select it in `SendEmailWithTopResources`. Replace `` in its From field with an address your mail server lets you send from. 4. Open `GetTopicFromToLearn` and press Execute workflow. The form opens in a new tab. Fill it in and submit. 5. Wait for the run, which can take minutes for a common topic, then read the email. ### How to use it Type a topic and your address and submit. The workflow searches Ask HN posts that match the topic, fetches the top-level comments under them one by one, joins them into one text and sends it to Gemini. The prompt asks for the best resources, grouped by type and level. A Markdown step turns the reply into HTML, and the email step sends it. What to change for a different result | To change | Where | | --- | --- | | The topic and the address | The two fields in \`GetTopicFromToLearn\` | | How many posts are searched | \`limit\` in \`SearchAskHN\`, 150 as shipped | | What the model looks for and how it lays out the list | The prompt in \`Basic LLM Chain\` | | The model | \`Google Gemini Chat Model\` | | The subject and wording of the email | \`SendEmailWithTopResources\` | ### Example A learner types `Rust` and an address and submits the form. The email lists resources under headings such as books, courses and videos, each with a one-line note and a link when a comment gave one. Treat it as a starting point. The model summarises what commenters said, so a resource can be dated or wrong. ### 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. This one starts from a form that n8n hosts for you. Hacker News is a discussion site, and the answers under an Ask HN post are often a good list of what people in the field trust. The workflow reads those answers so that you do not have to. **For experts.** The Hacker News node in `SearchAskHN` searches through Algolia. It returns up to 150 matching posts, and each post carries the ids of its top-level comments in a list called `children`. Next, `SplitOutChildrenIDs` makes one item per id. After that, `FindHNComments` fetches each comment from the Hacker News Firebase API. Then `CombineIntoSingleText` collects the comment texts into one list, and `Basic LLM Chain` places it in its prompt. Where to look closer | Node | What it does | What to know | | --- | --- | --- | | \`GetTopicFromToLearn\` | Shows the form and starts the run | The email field is required and the topic field is not | | \`SearchAskHN\` | Finds up to 150 Ask HN posts that match | A topic with few posts gives a thin list | | \`SplitOutChildrenIDs\` | Makes one item per comment id | A common topic gives thousands | | \`FindHNComments\` | Fetches each comment from the Hacker News API | One request per comment, in turn | | \`CombineIntoSingleText\` | Joins the comment texts | The comments are HTML | | \`Basic LLM Chain\` | Asks Gemini for the best resources | The prompt sets the categories and the layout | | \`SendEmailWithTopResources\` | Sends the HTML email | Needs an SMTP credential and a sender address | ### Good to know The workflow can read thousands of comments in one run, and its form is open to anyone who has the address. - Model: Google’s deprecations page, read on 12 November 2026, lists \`gemini-2.0-flash\` for shutdown on 1 June 2026 with \`gemini-3.6-flash\` as the replacement, and \`gemini-3.8-flash\` as the newest Flash model. The file’s \`gemini-1.5-flash\` is no longer listed. Mindket’s copy sets \`models/gemini-3.6-flash\` - Size: on 12 November 2026 the topic \`python\` returned 150 posts and 3,276 top-level comments. The run fetches each comment in turn and sends all the text to the model in one call, so it can take minutes and can pass the model’s input limit. Lower \`limit\` in \`SearchAskHN\` to cut both - Prompt fix: the prompt used a form field called \`Learn\` once, and the form has no such field, so that line came out empty. Mindket’s copy points it at the topic field - Sender: the file named its author’s own mailbox as the sender. It is now \`\\`, so fill in an address your SMTP account may send from - Judgement: the model weighs what commenters wrote. It may miss a good resource that nobody named, and it can repeat a claim that is wrong - Open form: once the workflow is active, anyone who has the form’s address can start a run and use your Gemini and mail quota - Your data: the topic goes to Hacker News’s search, the comment text goes to Google, and the address is used to send the email - Billing: one model call per run, sized by the comments collected - Credentials: the JSON holds none ## README ```markdown # Hacker News learning resources on any topic, sent by email An n8n workflow with a small form. You give it a topic and your email address. It collects the top-level comments under Ask HN posts on that topic, has a Google Gemini model pick out the books, courses, articles and videos people recommend, and emails you the list. ## What is inside - `workflow.json`: the workflow as n8n exports it, 10 nodes, no sticky notes. It holds no credential. - `README.md`: these notes. ## What we changed from the original copy - The model node and the email node were linked to credentials in their author's n8n. The links are gone, so select your own. - The model was `models/gemini-1.5-flash`, which Google no longer lists. It is now `models/gemini-3.6-flash`. - The sender address was the author's own mailbox. It is now ``. - The prompt used a form field called `Learn` once, and the form has no such field, so that line came out empty. It now reads the topic field, `I want to learn`. - The form's webhook id is gone, so n8n makes a new one. - Nothing else. The file has no Code node. ## Before you start - An n8n instance, on n8n Cloud or your own server. - A Google Gemini API key. - An SMTP account that can send mail: host, port, user and password. ## Set it up 1. In n8n, create a workflow, open its menu and choose Import from File. Pick `workflow.json`. 2. Add a Google Gemini (PaLM) API credential with your key. Select it in `Google Gemini Chat Model`. 3. Add an SMTP credential. Select it in `SendEmailWithTopResources` and replace `` in its From field with an address your mail server lets you send from. 4. Open `GetTopicFromToLearn` and press Execute workflow. The form opens in a new tab. Fill it in and submit. 5. Wait for the run. A common topic can take minutes. Then read the email. ## How a run goes 1. `GetTopicFromToLearn` shows a form with two fields, the topic and your email address, and starts a run when it is submitted. The page that follows tells the person an email is on its way. 2. `SearchAskHN` uses the Hacker News node to search for up to 150 posts tagged Ask HN that match the topic. 3. `SplitOutChildrenIDs` turns the list of comment ids on each post into one item per id. 4. `FindHNComments` fetches each comment from the Hacker News API, one request per id. 5. `CombineIntoSingleText` joins the `text` of all the comments into one list. 6. `Basic LLM Chain` puts that list in a prompt and sends it to `Google Gemini Chat Model`. The prompt asks for the best resources, grouped by type and by level, with a link when a comment gave one, and for the answer in Markdown. 7. `Convert2HTML` turns the Markdown into HTML. 8. `SendEmailWithTopResources` emails it to the address from the form, after a first line that says how many comments were read. 9. `Finished` marks the end. ## Change the look - The topic and the address: the two fields in `GetTopicFromToLearn`. - How many posts are searched: `limit` in `SearchAskHN`, 150 as shipped. - What the model looks for and how it lays the list out: the prompt in `Basic LLM Chain`. - The model: `Google Gemini Chat Model`. - The subject and the wording of the email: `SendEmailWithTopResources`. ## Good to know - Model: Google's deprecations page, read on 12 November 2026, lists `gemini-2.0-flash` for shutdown on 1 June 2026 with `gemini-3.6-flash` as the replacement, and `gemini-3.8-flash` as the newest Flash model. The file's `gemini-1.5-flash` is no longer listed. Mindket's copy sets `models/gemini-3.6-flash`. - Size: on 12 November 2026 the topic `python` returned 150 posts and 3,276 top-level comments. The run fetches each comment in turn and sends all the text to the model in one call, so it can take minutes and can pass the model's input limit. Lower `limit` in `SearchAskHN` to cut both. - The model judges what commenters wrote. It may miss a good resource that nobody named, and it can repeat a claim that is wrong. - The form is public once the workflow is active: anyone who has its address can start a run and use your Gemini and mail quota. - Your data: the topic goes to Hacker News's search, the comment text goes to Google, and the address is used to send the email. - Billing: one model call per run, sized by the comments collected. ``` ## What is in it 3 files, 13 KB, downloaded as a zip. ```text NOTICE.md 706 B README.md 4 KB workflow.json 8 KB ``` ## Workflow check Checked by Mindket on 13 Nov 2026. Read from the workflow’s file, not from a run. - Steps: 9 steps, starting on a form - AI: 1 step calls an AI model: Basic LLM Chain - Sign-ins: Signs in to: Email and Google Gemini - API keys: No API key is left in the file - Step types: Mindket’s checker reads all 9 of its step types ## Get it - [Get it on your Mindket account](https://account.mindket.com/account/?product=mindket#/market/66c91775-5061-4ba9-bb79-872a5f705f5a) - [Its page on Mindket](https://mindket.com/p/66c91775-5061-4ba9-bb79-872a5f705f5a)