A shareable config layer that turns Claude Code into a persistent personal AI with memory, tool integrations, and scheduled autonomous workflows.
khoj-ai/pipali
A desktop AI co-worker that reads and writes local files, runs code in a sandbox, and uses your browser to finish research and routine tasks.
- Stars
- 318
- Forks
- 61
- Open issues
- 5
- Last commit
- 26 days ago
What is it?
Pipali is a desktop AI co-worker for macOS, Windows and Linux. You assign it tasks such as researching across local documents and the web, drafting documents, spreadsheets and email, or running recurring workflows on a schedule. It works on your files and browser directly, so it can handle work that would otherwise take manual copying, searching and formatting.
Commands run inside a local sandbox that restricts file and network access. Anything requiring broader access needs your explicit approval, and you can adjust these permissions yourself. Custom workflows can be captured as skills based on the agentskills.io format, and external tools such as Jira, Linear and Slack can be connected via MCP.
Model access goes through the Pipali Platform with single sign-on, so no API key setup is needed. The platform offers several models, including Claude, GPT, Gemini, DeepSeek and Kimi.
Key features
- 01Runs commands in a local sandbox that restricts file and network access.
- 02Reads and writes files and operates the browser to complete tasks.
- 03Schedules recurring tasks or triggers them manually.
- 04Connects to Jira, Linear, Slack and other tools via MCP.
- 05Lets you define reusable workflows as skills.
- 06Provides model access through single sign-on without API keys.
Use cases
- Drafting weekly project update emails from your existing notes and tasks.
- Researching a topic across local documents and the web, then producing a brief.
- Automating monthly spreadsheet syncs or inbox cleanup rules.
- Creating Jira or Linear issues and Slack messages from a single request.
Quick start
- 1
Download the app for macOS, Windows or Linux from the download page
- 2
Sign in from the desktop app
- 3
Assign Pipali a task
Things to know
- The project is in Beta.
- Model access requires the Pipali Platform, and billing must be set up after initial signup credits run out.
Topics
FAQ
- What is pipali?
- A desktop AI co-worker that reads and writes local files, runs code in a sandbox, and uses your browser to finish research and routine tasks. Pipali is a desktop AI co-worker for macOS, Windows and Linux. You assign it tasks such as researching across local documents and the web, drafting documents, spreadsheets and email, or running recurring workflows on a schedule. It works on your files and browser directly, so it can handle work that would otherwise take manual copying, searching and formatting.
- Is pipali free to use? What license does it have?
- pipali is open source under the Apache-2.0 license. Check the license terms before using it in commercial products.
- Is pipali still maintained?
- Actively maintained. The last commit to pipali was on Sep 26, 2026.
- What can I use pipali for?
- Drafting weekly project update emails from your existing notes and tasks. Researching a topic across local documents and the web, then producing a brief. Automating monthly spreadsheet syncs or inbox cleanup rules. Creating Jira or Linear issues and Slack messages from a single request.
- How do I install pipali?
- Download the app for macOS, Windows or Linux from the download page Sign in from the desktop app Assign Pipali a task
- Where can I get pipali?
- The source code and installation instructions are on GitHub: https://github.com/khoj-ai/pipali
Summary generated by AI from the project's README. GitHub stats refresh daily.
Related projects
A local-first AI agent with one Rust core that works across terminal UI, desktop, mobile, CLI and chat apps.
An open source, general-purpose AI agent that runs on your machine as a desktop app, CLI, or embeddable API, for code and non-code tasks.
Open-source desktop AI agent by NetEase Youdao that runs office tasks on your local files and takes commands via IM apps like WeChat and Feishu.
A self-hosted AI agent runtime in .NET with a desktop companion app, supporting multiple model providers, tools, memory, and chat channels.
A private, local-first personal AI agent runtime that runs sandboxed tools, keeps memory and knowledge on your machine, and works with local, cloud, or Grok models.