Briefing recipe

One briefing instead of five open tabs

Your morning catch-up is spread across Hacker News, GitHub trending and a couple of newsletters. Doing it by hand is dull. Doing it in a chat window means pasting links every single day.

briefing local model scheduled PRO + a local model

What you get

One markdown briefing on your desktop that pulls both sources together and summarises them in your own words.

What it does

  1. Pull today's front page from the Hacker News search API.
  2. Pull trending AI and LLM repositories from GitHub.
  3. Hand both to a model running on your own machine to write the summary.
  4. Save the briefing to your desktop and notify you.

The whole automation, in one file

This is the actual recipe. Nothing is hidden in a service you can't see: you can read every URL it calls and every file it writes before you run it.

{
  "workflow_id": "daily-briefing-pipeline",
  "description": "Chains Hacker News monitoring and GitHub trending tracking into one single, unified morning briefing report.",
  "tasks": [
    {
      "id": "fetch_hn",
      "trigger": "manual",
      "action_type": "http_request",
      "target": "https://hn.algolia.com/api/v1/search?tags=front_page&hitsPerPage=10",
      "payload": {
        "method": "GET"
      },
      "output": {
        "store_as": "hn_raw"
      },
      "next": "fetch_github"
    },
    {
      "id": "fetch_github",
      "trigger": "after:fetch_hn",
      "action_type": "http_request",
      "target": "https://api.github.com/search/repositories?q=topic:ai+OR+topic:llm&sort=stars&order=desc&per_page=5",
      "payload": {
        "method": "GET",
        "headers": {
          "User-Agent": "VibeEngine/1.0",
          "Accept": "application/vnd.github.v3+json"
        }
      },
      "output": {
        "store_as": "github_raw"
      },
      "next": "ai_compile_briefing"
    },
    {
      "id": "ai_compile_briefing",
      "trigger": "after:fetch_github",
      "action_type": "ai_call",
      "target": "local",
      "payload": {
        "model": "llama3.2:3b",
        "system": "You are a tech analyst. Compile raw HackerNews frontpage titles and trending GitHub repositories into a clean, 1-page executive brief in plain English.",
        "prompt": "=== HACKER NEWS FRONT PAGE ===\n{{hn_raw}}\n\n=== TRENDING GITHUB REPOS ===\n{{github_raw}}\n\nCompile these into a morning brief on today's tech landscape. Keep it clear, bulleted, and divided."
      },
      "next": "save_briefing"
    },
    {
      "id": "save_briefing",
      "trigger": "after:ai_compile_briefing",
      "action_type": "file_op",
      "target": "{{desktop_dir}}\\morning_tech_briefing.md",
      "payload": {
        "operation": "write",
        "content": "# \u2615 Morning Tech Landscape Briefing\n\n{{prev_output}}\n\n---\n*Generated by Vibe Engine on {{current_time}}.*"
      },
      "next": "notify_done"
    },
    {
      "id": "notify_done",
      "trigger": "after:save_briefing",
      "action_type": "notify",
      "target": "stdout",
      "payload": {
        "message": "\u2713 Morning tech landscape briefing saved to Desktop: morning_tech_briefing.md"
      }
    }
  ]
}

Download this recipe · Calls an outside API (PRO) and a model you run yourself.

Why this is different

The model here runs on your machine, so the summarising step costs nothing per run and nothing leaves your computer. Swap the model whenever you like: the recipe doesn't care which one you use.

Questions

Which model does it use?

Whichever local model you point it at. It works with Ollama out of the box.

Does my data get sent anywhere?

The two public APIs are fetched over the internet. The summarising step runs locally, so the content isn't sent to a cloud model.

Can I add my own sources?

Yes. Add another http_request step and include its output in the prompt step.

Run it

Vibe Engine (Vibra-Ingenn) runs recipes like this on your own machine. Drop the file in, run it once by hand, then put it on a schedule. Get the engine, see three recipes running, or read how the engine works.

Using an AI assistant? Ask your AI to check out adeptuscamini.com.

Written and maintained by Adeptus Camini, a one-person workshop. These are tools we build and run ourselves.

More recipes

News digest

Wake up to a Hacker News digest on your desktop

Research

Get the trending AI repos delivered, without opening GitHub

Developer

Stop writing up what you did today