explaingit

holaboss-ai/holaos

5,657TypeScriptAudience · pm founderComplexity · 2/5LicenseSetup · easy

TLDR

holaOS is a macOS desktop app that turns repeating AI tasks into persistent workspaces where each review cycle saves your corrections as rules, making each run progressively more aligned with how you want the work done.

Mindmap

mindmap
  root((holaOS))
    What it does
      Recurring AI tasks
      Rule learning
      Parallel sub-agents
    Workspace features
      Goal and sources
      Output rules
      Run history
      Shared dashboard
    Use cases
      Weekly briefs
      Social media posts
      Customer feedback
      Launch campaigns
    Tech
      Electron app
      TypeScript
      macOS support
Click or tap to explore — scroll the page freely

Code map

Detail Auto

An interactive map of this repo's files and how they connect — its source is parsed live in your browser. Click Visualize to build it.

filefunction / class

Things people build with this

USE CASE 1

Set up a weekly competitive intelligence brief that automatically pulls sources, drafts a summary, and improves its writing style based on your edits each week.

USE CASE 2

Turn founder notes into draft social media posts and newsletter content, with your tone and style saved as rules from past corrections.

USE CASE 3

Aggregate customer feedback from multiple channels into a categorized product action board that updates every time you run the work-stream.

USE CASE 4

Manage a content launch campaign with parallel AI agents handling different tasks simultaneously inside an isolated workspace.

Tech stack

TypeScriptElectron

Getting it running

Difficulty · easy Time to first run · 30min

Currently macOS only in Beta 0.1, Windows and Linux support is listed as in progress.

The project uses a modified Apache 2.0 license, you can use and modify the software but the specific modifications may add additional restrictions beyond the standard Apache terms.

In plain English

holaOS is a desktop application that turns recurring, context-heavy work into what it calls AI work-streams. The idea is that many people re-open the same kind of task every week: a weekly research brief, a batch of social media posts, a review of customer feedback. Instead of starting from scratch in a chat window each time, holaOS lets you build a workspace around that repeating job. The workspace holds the goal, the relevant sources, the rules for how the output should look, and a history of past runs. When you review the output from a run and correct something, those corrections are saved as rules. The next run starts with those rules already in place, so the AI is not starting from zero. The README describes this as a loop where each review cycle makes the next output a little more accurate and a little more aligned with how you actually want the work done. Under the hood, holaOS runs on your local machine as an Electron desktop app, built in TypeScript. AI agents inside it can use the same browser, files, and apps that you use, rather than operating in an isolated sandbox. Complex tasks can be broken into parallel sub-tasks handled by multiple sub agents running at the same time. Each workspace also has a customizable dashboard that both you and the agent update as the work progresses. The README lists several example use cases the tool is built around: tracking competitors and shipping a weekly brief with source links, turning founder notes into draft social posts and newsletter content while saving your edits as style rules, grouping customer feedback from various channels into a product action board, managing launch campaigns with calendars and asset tracking, and running isolated workspaces for separate client projects. holaOS is currently in Beta 0.1 and officially supports macOS, with Windows and Linux support in progress. It installs via a single curl command. The license is a modified version of the Apache 2.0 license.

Copy-paste prompts

Prompt 1
I want to build a holaOS work-stream that monitors 5 competitor websites weekly and generates a brief with their product updates. Walk me through setting up the workspace, sources, and output rules.
Prompt 2
Show me how to create a holaOS rule that captures my editing style so the AI matches my newsletter tone without me re-explaining it on each run.
Prompt 3
I have a holaOS workspace for customer feedback analysis but the output categories keep changing between runs. How do I lock in a consistent output structure using rules?
Prompt 4
Walk me through installing holaOS on macOS using the curl command, then creating my first work-stream for drafting weekly social media posts from bullet-point notes.
Open on GitHub → Explain another repo

← holaboss-ai on gitmyhub — every repo by this author, as a profile.

Verify against the repo before relying on details.