PrimeIntellect-ai/prime-agent
Self-improving RLM agent for coding and long-running tasks
- Chart
- Not on a chart right now
- Stars
- 22.8k
What it is
Prime Agent is an open-source coding and research agent harness built around a Recursive Language Model (RLM) and a Continual Harness. It runs inside a persistent Python REPL and stores supplemental prompts, memories, and skill descriptions as durable state.
Who it is for
It is suited for developers and researchers running long-running coding workflows or autonomous research evaluations.
Why it is trending
It reached 22,798 total stars with 67 gained today and released v0.10.3 on November 12, 2026. It stands out for daemon-backed session continuity, direct agent-to-agent messaging, and rollback-safe harness refinement via `/refine`.
Main features
- Programmatic file, shell, and tool execution inside a persistent Python REPL
- Built-in subagent spawning with `rlm.spawn(...)` and direct agent messaging
- Evidence-backed updates to harness state with `/refine` and snapshot rollback
- Daemon-backed sessions, `/goal` tracking, schedules, and bounded `/autonomous` mode
Summarized on November 17, 2026 from the README and repository data. It can be wrong; check the source before relying on it. Report a wrong summary