Skip to content

Entry

cua

Appears in 4 awesome lists

Open-source infrastructure for Computer-Use Agents: sandboxed full-desktop control across macOS, Linux, and Windows, a background-native macOS driver that operates without stealing cursor focus, plus SDKs and benchmarks. The most complete reference for building harnesses around screen-based agent…

Open github.comtrycua/cua

Found in these lists

Awesome Harness Engineering

Section: Demo Harnesses · Open-source infrastructure for Computer-Use Agents: sandboxed full-desktop control across macOS, Linux, and Windows, a background-native macOS driver that operates without stealing cursor focus, plus SDKs and benchmarks. The most complete reference for building harnesses around screen-based agent…

FreshScore 93

Awesome local LLM

Section: Computer Use · the Docker Container for Computer-Use AI Agents

FreshScore 90

Awesome Open Source AI

Section: 4. Agentic AI & Multi-Agent Systems · Open-source sandboxes, SDKs, and benchmarks for computer-use agents to control desktop environments. MIT licensed.

FreshScore 87

Awesome Jev

Section: Browser & computer use (77) · Computer-use platform with a Jev recipe where the model selects one bounded action candidate and the client validates it before execution.

FreshScore 85

LangChain

Langchain integrates various providers like Anthropic, AWS, and OpenAI, and offers tools for components such as LLMs, chat models, and data analysis, supporting functionalities from Alpha Vantage to YouTube github | docs

In 22 listsDetails

mem0

The most-deployed memory layer for AI agents — drop-in add/search/update/delete fact storage with LLM-extracted long-term / short-term / semantic memory: automatic fact consolidation (dedup, contradiction resolution, decay) so agents keep "context that persists" without context bloat;…

In 14 listsDetails

Dify

February 2026 release making human oversight a native workflow primitive: suspend execution at critical decision points, expose review-and-edit UI mid-flow, and route subsequent execution based on human action (approve/reject/escalate). Demonstrates how HITL transitions from bolt-on approval gates…

In 14 listsDetails

PydanticAI

June 2026 harness-first redesign built around the Capability primitive: a single composable unit bundling instructions, tools, lifecycle hooks, and model settings. The split between a small stable core and a fast-moving pydantic-ai-harness lets capabilities graduate as they prove essential, while…

In 13 listsDetails

anthropics/skills

by Anthropic - Anthropic's official repository for Agent Skills — the SKILL.md format, a skill template, and example skills, the same format Claude Code loads natively.

In 12 listsDetails

Jev Social

Read-only Instagram, TikTok and LinkedIn research app where Jev selects bounded operations, the local socai CLI runs them in the user's Chrome, and captured evidence becomes cited reports; TikTok media is optional and explicit-request only.

In 12 listsDetails

AutoGen

Microsoft's multi-agent conversation framework with a complete AgentChat layer covering agent loop, tool integration, termination conditions, and human-in-the-loop. The most comprehensive open-source reference for large-scale multi-agent harness design.

In 14 listsDetails

browser-use

Open-source browser agent (Python and TypeScript) whose loop injects serialized DOM state each step, exposes browser actions as registered tools, and runs the same code against a local Chrome or a cloud browser. It has grown from a minimal reference into a full framework (~70k lines of Python…

In 11 listsDetails