You don't need a chat ui but a canvas with ai.
[ACL 2026] LLM-Based Human-Agent Collaboration and Interaction Systems: A Survey | Awesome Human-Agent Collaboration | Human-AI Collaboration
Framework and toolkits for building and evaluating collaborative agents that can work together with humans.
Local-first coordination and authority layer for long-running human-agent projects
Spring Voyage — A human-agent collaboration platform
SYMBIVELA - Human-Agent Collaboration Workspace. Set goals, review machine plans, approve high-risk actions, resolve exceptions, intervene in execution, accept outcomes, and inspect evidence across agents, workflows, robots, and twins.
Team templates for human × agent collaboration — dev, sales, marketing, and more. Powered by hxa-connect.
An open source protocol for Human Agent Interaction
internOS — Workstreams framework for human-agent collaboration. Coordinate work across projects, tasks, communication threads, and filesystem workstreams for any AI agent framework.
CyberAgent — agent orchestration & human-agent collaboration: agentic project and task management where coding agents and people ship work together, with review gates, shared context and full audit. By Hyperspace.
Embodied agent that learns from human teaching; hybrid agent architecture with computer vision, AI planning, knowledge graph reasoning & generalization , inverse kinematics. Supported by the DARPA GAILA program.
A conversation-card deck for exploring, challenging, planning, and explaining ideas with AI.
人体智能体协作模式-多智能体架构系统
A declarative specification language for human-agent collaboration. Define what and what constraints, not how. Built with Rust + PEG grammar (pest). 🤖 Co-developed with Claude (Anthropic).
The semantic protocol behind Control Decks. Cards are the interface. Control is the protocol.
Co-Gym extension for evaluating collaborative agents under changing human-specified delegation.
Self-describing message protocol with immutable, replayable history for human-agent collaboration. Types and authority rules live in the append-only log—state reconstructed from cryptographically sealed messages.
A knowledge coordination protocol for human-agent collaboration. Connects research → strategy → requirements → implementation into a traversable, version-aware graph.
Engineering discipline and reliability for frontier coding agents: durable responsibility, adversarial verification, continuity, and autonomous completion.
Local deterministic CLI that assembles inspectable receipts for AI coding-agent work sessions from declared artifacts.
To associate your repository with the human-agent-collaboration topic, visit your repo's landing page and select "manage topics."