DeerFlow
Open-source super agent harness by ByteDance for long-horizon research, coding, and creation with sub-agents, memory, sandboxes, and skills.
🛡️ AgentReady threat assessment
MAESTRO 7-layer threat model + OWASP AIVSS risk score for DeerFlow, derived from its capabilities.
These scores are auto-generated from public information (the agent's own listing, docs, and repository) using the canonical OWASP AIVSS formula and the MAESTRO framework — an estimate for guidance, not a penetration test, audit, or certification. See the scoring methodology — every score is re-derived by the same automated method as an agent's public evidence changes.
Overview
DeerFlow, from ByteDance, is an open-source long-horizon super agent harness for tasks such as research, coding, and content creation. The GitHub repository describes DeerFlow as orchestrating sub-agents, memory, sandboxes, tools, skills, and a message gateway to handle tasks that can take minutes to hours. DeerFlow 2.0 is a ground-up rewrite of the original Deep Research framework, with the earlier framework maintained on the 1.x branch. The project is distributed on GitHub under the MIT license.
Key features and capabilities
- open source
- super agent harness
- multi-agent orchestration
- long-horizon tasks
- research automation
- coding automation
- sandboxed execution
- memory
- sub-agents
- skills
Use cases
- Building long-horizon autonomous workflows
- Coordinating sub-agents for complex tasks
- Research automation and deep exploration workflows
- Coding and software task automation
- Experimenting with memory, sandboxes, tools, and extensible skills