PulseAugur
EN
LIVE 22:14:37

Andrew Ng releases OpenWorker, a local AI agent for finished deliverables

Andrew Ng has launched OpenWorker, an open-source, local-first desktop AI agent designed to complete tasks and produce finished deliverables rather than just engaging in conversation. The agent operates entirely on the user's machine, supporting a wide array of models including those from OpenAI, Anthropic, Google, and local Ollama instances. OpenWorker distinguishes itself with a robust permission engine that classifies tool calls by risk and allows users to define custom approval modes, ensuring user control over consequential actions. AI

IMPACT This tool could streamline workflows by automating task completion, potentially reducing the need for direct conversational interaction with AI for specific outcomes.

RANK_REASON This is a new open-source tool release from a prominent figure in AI, but not a frontier model release from a major lab.

Read on Mastodon — fosstodon.org →

AI-generated summary · Google Gemini · from 2 sources. How we write summaries →

Andrew Ng releases OpenWorker, a local AI agent for finished deliverables

COVERAGE [2]

  1. MarkTechPost TIER_1 English(EN) · Asif Razzaq ·

    Andrew Ng Just Released OpenWorker: An Open-Source, Local-First Desktop AI Coworker That Returns Finished Deliverables Instead of Chat

    <p>Andrew Ng has released OpenWorker, an MIT-licensed desktop AI agent that returns finished deliverables instead of chat replies. It runs a local Python agent server under a Tauri shell, supports 30 curated tool-calling models plus fully local Ollama, and gates every write, shel…

  2. Mastodon — fosstodon.org TIER_1 English(EN) · [email protected] ·

    Andrew Ng has released OpenWorker, an open-source desktop AI agent that produces finished work rather than conversation. It runs locally, supports 30+ tool-call

    Andrew Ng has released OpenWorker, an open-source desktop AI agent that produces finished work rather than conversation. It runs locally, supports 30+ tool-calling models plus Ollama, and gates all write and shell actions behind a typed risk engine. https://www. marktechpost.com/…