# OpenManus Open-source framework for a general-purpose agent that you install yourself and start in the terminal with your own model key. Canonical profile: https://ki-agentenkompass.de/en/agents/openmanus/ Editor: [Michael Meierhoff](https://ki-agentenkompass.de/en/about/) Sources reviewed: 2026-11-05 Basis: provider documentation. Examples are editorial suggestions, not hands-on product tests. ## Overview OpenManus is a community project of the GitHub organisation FoundationAgents. The README calls it a simple implementation and says Manus can be tried without an invite code; core authors are from MetaGPT, according to the README. The latest commit on the main branch dates from 16 August 2026 (GitHub API, checked on 5 November 2026). The repository is not archived, but development is quiet: the latest release, v0.3.0, is from 10 April 2025. - Provider: FoundationAgents - Product family: OpenManus - Profile type: Open-source agent framework - Environment: Python 3.12 on your own machine; input in the terminal - Connections: Model providers per the example configuration, MCP, Browser Use as browser tool, optional Docker sandbox - Setup: Clone the repository, install dependencies with conda or uv, create config.toml from the example file and enter a model key. - Pricing model: Open source under the MIT licence. Model access needs your own API key or a local model. ## Key features ### General-purpose agent in the terminal You start the agent with python main.py and enter your idea in the terminal. Source: https://github.com/FoundationAgents/OpenManus ### Browser automation through Browser Use OpenManus starts Browser Use CLI 3.0 as an MCP server. Locally it attaches to Chrome or Chromium without an API key; a cloud browser needs a Browser Use key. OPENMANUS_DISABLE_BROWSER_USE=1 turns it off. Source: https://github.com/FoundationAgents/OpenManus ### MCP version python run_mcp.py starts the variant with MCP tools. Source: https://github.com/FoundationAgents/OpenManus ### Multiple agents in flow mode python run_flow.py starts the multi-agent version, which the README calls unstable. A data analysis agent is off by default and can be enabled in config.toml. Source: https://github.com/FoundationAgents/OpenManus ### Configurable models The example configuration contains entries for Anthropic, Amazon Bedrock, Azure OpenAI and Ollama, among others, plus an optional vision model. Source: https://raw.githubusercontent.com/FoundationAgents/OpenManus/main/config/config.example.toml ### Optional Docker sandbox The example configuration has a commented-out [sandbox] section with Docker image, memory and CPU limits, timeout and network switch; use_sandbox is set to false there. Source: https://raw.githubusercontent.com/FoundationAgents/OpenManus/main/config/config.example.toml ### Selectable search engine According to the example configuration, the search engine can be set to Google, Baidu, DuckDuckGo or Bing, with a fallback order. Source: https://raw.githubusercontent.com/FoundationAgents/OpenManus/main/config/config.example.toml ## Example workflows ### Research idea in the terminal - Your input: A question or idea you type into the terminal after python main.py. - Possible result: The agent works through the task with the configured model and its tools. - What to check: Check sources, page visits and results yourself, especially with the browser tool active. ### Data analysis in flow mode - Your input: A file or task to analyse, data analysis agent enabled in config.toml. - Possible result: Analysis and charts through run_flow.py, which the README calls unstable. - What to check: Compare numbers and charts with the raw data and install the extra dependencies. ## Requirements and limits ### Python 3.12 and your own model key Requirement: Both install routes in the README require Python 3.12. Without an entry in config.toml the agent has no model. Source: https://github.com/FoundationAgents/OpenManus ### Key in the configuration file Limitation: The README has you enter the API key directly in config/config.toml. Source: https://github.com/FoundationAgents/OpenManus ### Flow mode unstable Limitation: The README calls run_flow.py the unstable multi-agent version. Source: https://github.com/FoundationAgents/OpenManus ### Sandbox not active by default Limitation: In the example configuration the sandbox section is commented out and use_sandbox is false. Source: https://raw.githubusercontent.com/FoundationAgents/OpenManus/main/config/config.example.toml ### Privacy, schedules and memory Not yet confirmed: The README describes no privacy statement, no schedules and no memory across sessions. Source: https://github.com/FoundationAgents/OpenManus ## A good fit if … - you want to run an open-source general-purpose agent yourself and try it with your own model key - you want to plug Browser Use, MCP or your own tools into a Python agent - you want to read and adapt the code and can live with a quiet pace of development ## Less suitable if … - you expect a hosted product with an account, support and administration - you need documented memory, schedules or approval per action - you need stable multi-agent operation without experiments Editorial assessment based on provider documentation. Not a product test. ## Usable AI models Open choice You enter the model, base URL and key in the configuration yourself. The example file shows Anthropic, OpenAI via Azure, Bedrock, the Jiekou.AI gateway and Llama locally via Ollama. No model access is included, so you need your own account or a local server. - Model families named: Anthropic (Claude), OpenAI (GPT), Meta (Llama, Muse) · Other models can be connected freely - Model choice: You choose - Your own API key: Yes - Router or custom endpoint: Yes - Local models: Yes Documented · Sources accessed on: 2026-11-05 - [OpenManus: config.example.toml](https://raw.githubusercontent.com/FoundationAgents/OpenManus/main/config/config.example.toml) - [GitHub: OpenManus](https://github.com/FoundationAgents/OpenManus) ## Europe & regional availability README, example configuration and project page reviewed. They name no country release or data region. Reviewed scope: Local software; model providers separately Reviewed on: 2026-11-05 ### Europe: no statement from the project Availability still unclear OpenManus runs on your own machine. The reviewed pages say nothing about Europe. What data goes to model providers or to the optional Browser Use cloud browser depends on your configuration; the README does not describe this. Region covered: EU/EEA, United Kingdom and Switzerland Documented reason: The reviewed sources name no country release for this. Official source: https://github.com/FoundationAgents/OpenManus Official source: https://raw.githubusercontent.com/FoundationAgents/OpenManus/main/config/config.example.toml ## Accounts & permissions Based on provider documentation. Permissions depend on the account, integration and organization policies. Review the permissions actually requested before use. Configuration covered: Self-run agent with your own model key Sources reviewed: 2026-11-05 ### Account used Documented: You enter the model key directly in config/config.toml. A cloud browser additionally needs a Browser Use key as an environment variable. Evidence: https://github.com/FoundationAgents/OpenManus ### Admin approval Open in reviewed sources: The reviewed pages describe no approval or rights management by administrators. Evidence: https://github.com/FoundationAgents/OpenManus Evidence: https://openmanus.github.io/ ### Data access & team sharing Documented with conditions: Locally, the browser tool attaches to Chrome or Chromium. The example configuration has a commented-out sandbox section with use_sandbox = false. Evidence: https://github.com/FoundationAgents/OpenManus Evidence: https://raw.githubusercontent.com/FoundationAgents/OpenManus/main/config/config.example.toml ## Prices and included usage Pricing checked: 2026-11-05 OpenManus is under the MIT licence; the reviewed pages name no price. Costs arise for model access with the chosen provider. ### OpenManus No software fee stated Agent framework with your own model access. Price source: https://raw.githubusercontent.com/FoundationAgents/OpenManus/main/LICENSE The optional Browser Use cloud browser needs its own key according to the README; terms were not reviewed. Pricing details: https://github.com/FoundationAgents/OpenManus ## Operation & oversight Based on public provider documentation. Assessments are editorial inferences from the stated properties. Actual time requirements and reliability have not been measured. Deployment covered: Local operation in the terminal Sources reviewed: 2026-11-05 ### Files and working state Open in reviewed sources: The README does not describe whether or where session state is kept. Evidence: https://github.com/FoundationAgents/OpenManus ### Sign-ins Documented: Access is a model key in config.toml; for the cloud browser, a Browser Use key. The README names no time limits. Evidence: https://github.com/FoundationAgents/OpenManus Evidence: https://raw.githubusercontent.com/FoundationAgents/OpenManus/main/config/config.example.toml ### Starting further runs Documented: You start runs yourself with python main.py, run_mcp.py or run_flow.py. The README describes no schedules or triggers. Evidence: https://github.com/FoundationAgents/OpenManus ### Human intervention Open in reviewed sources: The README describes no confirmation before actions. Input happens in the terminal. Evidence: https://github.com/FoundationAgents/OpenManus ### Detecting errors Open in reviewed sources: The README says nothing on error handling and retries. The example configuration mentions retries only for search engines (delay and maximum count). Evidence: https://github.com/FoundationAgents/OpenManus Evidence: https://raw.githubusercontent.com/FoundationAgents/OpenManus/main/config/config.example.toml ### Resuming after failures Open in reviewed sources: The README does not describe how to continue interrupted work. Evidence: https://github.com/FoundationAgents/OpenManus ### Editorial assessment You run and supervise OpenManus yourself. The reviewed pages describe no approvals, memory or error handling; plan your own isolation, for example through the optional sandbox. ## Memory and conversation continuity Scope: Self-run agent framework Sources accessed: 2026-11-05 The reviewed pages describe no memory and no stored sessions. ### Storage and scope Not publicly established: The README describes no storage location for memory or sessions. The example configuration mentions only memory limits for the optional sandbox. - [OpenManus README](https://github.com/FoundationAgents/OpenManus) - [OpenManus example configuration](https://raw.githubusercontent.com/FoundationAgents/OpenManus/main/config/config.example.toml) ### Selection and timing Not publicly established: Not described: what the agent keeps or when. - [OpenManus README](https://github.com/FoundationAgents/OpenManus) ### Inspect, edit and delete Not publicly established: Not described: how you view, edit or delete content. - [OpenManus README](https://github.com/FoundationAgents/OpenManus) ### Agent and channel boundaries Not publicly established: Not described. The README names the launch variants main.py, run_mcp.py and run_flow.py, but no memory boundaries between them. - [OpenManus README](https://github.com/FoundationAgents/OpenManus) ### Retrieval Not publicly established: Not described. - [OpenManus README](https://github.com/FoundationAgents/OpenManus) ### Session reset and compaction Not publicly established: Not described; the README names neither session end nor compaction. - [OpenManus README](https://github.com/FoundationAgents/OpenManus) ## Context Your own machine with Python 3.12 and a model key that you enter directly in config.toml. The multi-agent Flow mode is described as unstable in the README, and the sandbox section is commented out in the example configuration. README, example configuration, licence and project page reviewed. Commit date, release and stars come from the GitHub API. Example tasks are editorial; hands-on tests are still pending. ## Open points Europe, availability still unclear: - Europe: no statement from the project Requirements and limits, not yet confirmed: - Privacy, schedules and memory ## Sources - https://github.com/FoundationAgents/OpenManus - https://openmanus.github.io/ - https://raw.githubusercontent.com/FoundationAgents/OpenManus/main/config/config.example.toml - https://raw.githubusercontent.com/FoundationAgents/OpenManus/main/LICENSE ## Provider Open-source community project on GitHub under the MIT licence. The project page names no operators; the README points to a contact within the project. - Operator: FoundationAgents - Organization: Provider - Website: https://openmanus.github.io/ - Reviewed: 2026-11-05 - Repository: https://github.com/FoundationAgents/OpenManus - GitHub-Stars: 58891 (2026-11-05) - License: MIT - Provider source: https://openmanus.github.io/ - Provider source: https://github.com/FoundationAgents/OpenManus - Provider source: https://raw.githubusercontent.com/FoundationAgents/OpenManus/main/LICENSE [Sources and method](https://ki-agentenkompass.de/en/sources/)