Privacy Policy

Privacy Policy

Learn about how OpenFang handles website access data, self-hosted deployment data, Agent OS runtime data, LLM APIs, communication channels, and the boundaries of user control.

Last updated: July 10, 2026

Who We Are

OpenFang is an open-source, production-grade Agent OS project built with Rust. This website serves to introduce OpenFang's features, Hands, installation methods, performance benchmarks, documentation, and related open-source resources. OpenFang is not a cloud-hosted chat service, nor is it an official product of Telegram, Discord, Slack, Feishu, DingTalk, or any LLM service provider.

Information Collected by the Official Website

When you visit openfang.cc or our test domains, our servers, security services, or hosting providers may record basic access logs, such as request timestamps, pages visited, browser type, IP addresses, error logs, and security events. This information is used to maintain website availability, troubleshoot issues, prevent abuse, and understand site traffic.

Cookies and Basic Preferences

This website may use essential cookies to save language settings, debugging access, or basic session preferences. Unless explicitly stated on a page, we do not require you to submit LLM API Keys, channel tokens, business workflows, Agent memory, private code, or production data through this website.

Data in Self-Hosted Deployments

If you self-host OpenFang, your runtime data is controlled entirely by your own devices, servers, file systems, and third-party accounts. This may include Hands workflows, channel adapter tokens, message integrations for platforms like Telegram, Discord, Slack, Feishu, and DingTalk, as well as audit logs, task records, configuration files, Agent memory, LLM API Keys, and deployment logs.

Boundaries Between the Website and Deployment Environments

This self-hosted runtime data is not automatically sent to the OpenFang official website simply because you visit it. You are responsible for securing your deployment environment, access permissions, keys, backups, log retention policies, audit chains, and API Key security.

Third-Party Services

OpenFang can connect to LLM providers such as Anthropic, OpenAI, Gemini, Groq, DeepSeek, Mistral, Together, Ollama, and vLLM, and may interact with platforms like Telegram, Discord, Slack, Feishu, DingTalk, GitHub, or various hosting and analytics services. When using these services, you must read and comply with the respective third-party privacy policies and terms of service.

User Control and Security Recommendations

Please do not submit API Keys, channel tokens, private configurations, business workflows, customer data, or sensitive audit logs to public issues, chat groups, or unprotected logs. We recommend using the principle of least privilege, isolated deployments, regular dependency updates, and restricting access to your OpenFang instance to trusted users only.

Children's Privacy and Policy Updates

OpenFang is intended for users capable of self-deploying and managing software, workflows, and Agents; it is not intended for children. We may update this policy as the website, project features, or third-party dependencies change. Updates will be reflected with a new date on this page.

Contact Us

If you have questions regarding this privacy policy or the OpenFang project, please contact the maintainers via the project's public GitHub repository or community links. Please do not submit sensitive credentials or private data through public channels.

View Project Source Code

For full licensing, installation instructions, and technical details, please refer to the GitHub repository.

Open GitHub