Agent Orchestrator

Agent Orchestrator (AO) is an open-source multi-agent harness developed by Untrivial-ai. Built to coordinate and execute autonomous development workflows, it provides a orchestration environment designed to run and supervise terminal-based AI coding agents across complex engineering tasks. In day-to-day operation, Agent Orchestrator manages agent lifecycles and background execution sessions, leveraging tmux session management to isolate agent states and run processes concurrently. It natively supports configurable skill directories—such as `.agents/skills` and `.claude/skills`—allowing teams to plug in agent capabilities, bug triage workflows, and custom tool interfaces directly into terminal development sessions. Agent Orchestrator differentiates itself by functioning as a session-aware harness that coordinates persistent agent execution environments rather than operating as a standalone interactive chat prompt. By managing state across sessions and decoupling agent logic through modular skill definitions, it enables parallelized task distribution and multi-agent execution in local development workflows. Agent Orchestrator is distributed as an open-source project on GitHub under standard community licensing, with an active repository structure supporting CI-driven security checks and developer workflows.

Tags: agent harness, ai agent, cli, coding agent, developer tools, OpenSource

Visit Agent Orchestrator

Key features

  • Tmux-based session management — Isolates agent execution states within dedicated tmux sessions for concurrent and persistent task execution. - Modular skill directory support — Discovers and loads custom tool definitions and prompts from local .agents/skills and .claude/skills directories. - Automated bug triage workflows — Incorporates dedicated triage and issue processing skills to investigate and address repository bugs. - Multi-agent orchestration — Coordinates multiple agent instances to divide and conquer complex repository-level code modifications. - CI/CD and Git integration — Integrates with GitHub workflows, pull request handling, and automated security scanning.

Use cases

Use Case: Concurrent Multi-Agent Development A developer delegates multiple independent refactoring or documentation tasks to separate orchestrated agents, running them concurrently across isolated tmux sessions without process collision.

Use Case: Repository Bug Triage A team configures triage skills to automatically inspect incoming bug reports, clone the required context, reproduce issues, and formulate pull requests directly from the CLI.

Use Case: Custom Skill Execution for Claude Agents An engineer defines project-specific tools in .claude/skills and uses Agent Orchestrator to execute structured multi-step automation tasks against local codebases.

Who it is for

Agent Orchestrator is designed for software engineers and DevOps teams looking to automate complex repository tasks with multi-agent orchestration and terminal-based isolation. It is less suitable for developers who only want a lightweight, single-file inline completion tool or IDE autocomplete plugin.

Install guide

Installation Clone the repository directly from GitHub and install project dependencies (standard repository setup): bash git clone https://github.com/Untrivial-ai/agent-orchestrator.git cd agent-orchestrator

First Run Ensure tmux is installed on your system to support session orchestration, then configure any required agent API credentials: bash # Ensure tmux is available tmux -V # Configure API keys for underlying agent backends (e.g., Anthropic or OpenAI) export ANTHROPIC_API_KEY="your-anthropic-api-key"

Example Session Run the orchestrator against a repository to trigger skill-based execution or triage routines: bash # Launch an orchestrated session with local skills enabled ao run --skills-dir .claude/skills "Triage issue #4174 and suggest a pull request fix"

Part of CLI Coding Agents

Categories: AI Coding Agents, Code Assistants, Developer Tools

Related CLI agents

  • SWE-agent — SWE-agent is an open-source autonomous software engineering agent harness built and maintained by researchers from Princeton University and Stanford University.…
  • Agent Runbook — Agent Runbook is an open-source contract-based multi-agent skill framework developed by KnoxOps. It acts as an orchestrator and compiler that…
  • Weaver — Weaver is an open-source agent harness and orchestration framework maintained by sean35mm on GitHub. Built on top of the Bun…
  • Context Bridge — Context Bridge is an open-source CLI agent harness and context orchestration utility developed by serdardb. It is built to facilitate…
  • Hivelore — Hivelore is an open-source agent harness and contextual memory orchestrator created by Doucs91. It provides persistent project memory, decision tracking,…
  • Agentic Engineering Framework — Agentic Engineering Framework is an open-source orchestration harness and governance layer developed by Dimitri Geelen. It is designed to constrain,…
  • AI Tools
  • Categories
  • Industries
  • CLI Coding Agents
  • MCP Servers
  • MCP Categories