QwenLM/qwen-code
Qwen Code 1.0 Preview packages memory, skills, subagents, teams, and MCP in an open-source coding agent with multiple provider and client options.
**Qwen Code 1.0 Preview** combines Qoder and Qwen Code harness capabilities, including Auto-Memory, Auto-Skills, SubAgents, Agent Teams, and MCP. It supports OpenAI, Anthropic, Gemini, and Qwen APIs, plus terminal, desktop, web, IDE, SDK, and chat interfaces. Builders can try it as a multi-provider agent or use its headless mode and SDKs for automation. Its daemon can share sessions across clients and workspaces, with HTTP and SSE access. The daemon listens on loopback without bearer authentication by default, and local processes can access its API, including code execution. The project recommends setting a token on shared or untrusted hosts; daemon mode is marked experimental.
No separate implication has been recorded yet.
**Qwen Code 1.0 Preview** combines Qoder and Qwen Code harness capabilities, including Auto-Memory, Auto-Skills, SubAgents, Agent Teams, and MCP. It supports OpenAI, Anthropic, Gemini, and Qwen APIs, plus terminal, desktop, web, IDE, SDK, and chat interfaces. Builders can try it as a multi-provider agent or use its headless mode and SDKs for automation. Its daemon can share sessions across clients and workspaces, with HTTP and SSE access. The daemon listens on loopback without bearer authentication by default, and local processes can access its API, including code execution. The project recommends setting a token on shared or untrusted hosts; daemon mode is marked experimental.
Qwen Code 1.0 Preview combines multi-provider agent features and many client surfaces with a daemon for shared sessions and automation. The daemon's unauthenticated loopback default is a consequential deployment constraint: local processes can reach its API, including code execution, and shared or untrusted hosts need a token; daemon mode remains experimental.