Paseo

Paseo is an open-source AI agent harness and orchestration framework maintained by the getpaseo project. Designed for managing terminal-based and automated coding workflows, Paseo structures agent execution environments through modular packages, extensible skills, and dedicated execution runtimes. In daily use, Paseo acts as a bridge between LLM providers and developer environments. It uses a structured skills system (supporting standardized skill definitions in directories like `.agents/skills` and `.claude/skills`) and a plugin architecture to provide AI agents with fine-grained access to local tools, repositories, and execution contexts. Developers can configure agent behavior and constraints via project-level configuration files. Paseo differentiates itself through its strong emphasis on reproducibility and containerized runtime support. With built-in Nix environments, Docker configuration, and a modular monorepo architecture, Paseo provides predictable execution boundaries for autonomous agents, preventing uncontrolled side effects on host machines. Paseo is distributed as open-source software under an open license on GitHub. It is actively maintained with support for standard package managers, multi-runtime configurations via Mise, and modern linting infrastructure.

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

Visit Paseo

Key features

  • Extensible skills system — Organizes agent tools and instructions via dedicated skill directories such as .agents/skills and .claude/skills. - Plugin ecosystem — Supports custom plugins and extensions to connect external APIs, developer tools, and workflow automations. - Containerized execution — Includes preconfigured Docker setups to isolate agent actions and run commands safely. - Nix environment support — Provides Nix flakes and shell definitions for reproducible developer environments across different platforms. - Monorepo architecture — Designed as a modular set of packages allowing integration as a standalone CLI or embedded orchestrator. - Standardized agent contracts — Uses project-level specification files like AGENTS.md and CLAUDE.md to define operational constraints.

Use cases

Use Case: Sandboxed Task Execution Developers can run autonomous coding agents inside Paseo's Docker or Nix environments to refactor code or run tests without granting unconstrained access to the host system.

Use Case: Custom Skill and Tool Integration Teams building domain-specific workflows can author modular skills in .agents/skills to expose internal CLI utilities and custom API endpoints directly to LLM agents.

Use Case: Cross-Agent Workflow Orchestration Engineers can standardize multi-agent instructions across repositories using shared configuration files like AGENTS.md, ensuring consistent behavior regardless of the underlying LLM provider.

Who it is for

Paseo is targeted at software engineers, DevOps teams, and platform developers who need a customizable, reproducible harness for orchestrating autonomous coding agents and custom toolsets. It is less suited for developers seeking an out-of-the-box GUI coding assistant with no configuration requirements.

Install guide

Installation Paseo can be cloned and built directly from source using standard Node/TypeScript tooling or Mise: bash # Clone the repository git clone https://github.com/getpaseo/paseo.git cd paseo # Install dependencies and build packages npm install npm run build Alternatively, if using Nix for isolated builds: bash nix develop

First Run Configure your environment variables and LLM credentials before starting the orchestrator: bash export ANTHROPIC_API_KEY="your-api-key" # Configure optional plugin and tool settings in .env or .mise.toml

Example Session Run the agent against a repository using configured skills: bash # Execute a task using defined project skills paseo run "Inspect failing tests in packages/core and generate a patch"

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