PrimeIntellect-ai/prime-agent
Summary
Prime Agent is an open-source coding and research agent designed for general and long-running work. It centers on two abstractions: a Recursive Language Model (RLM) that treats context as variables and tools like subagents as function calls, and the Continual Harness that stores prompts, memories, skills, and reusable subagent specifications for durable state. The project combines a persistent Python control environment with durable harness state to enable persistent sessions, inter-agent communication, and autonomous long-running tasks, with guidance on installation and usage.