RSIHub: Self-Evolving Agent Framework
RSIHub is an open-source framework by Simple Agent Lab for building self-evolving agents. It enables evaluator-driven learning, reproducible candidate lineage, and controllable modification.
About this tool
Overview
RSIHub is an open-source framework for building agents that improve themselves through evaluator-driven learning. It emphasizes reproducible candidate lineage and controllable modification, allowing developers to track agent evolution and preserve evidence.
Key Capabilities
- Evaluator-driven learning: Agents train and improve through evaluator feedback.
- Reproducible candidate lineage: Records each modified candidate version for traceability.
- Controllable modification: Limits what agents can modify via declared surface and editable_roots.
- Multi-agent support: Supports codex, claude-code, aider, gemini-cli, mini-swe-agent, opencode, openhands, swe-agent, and others.
- Workspace management: Creates new workspaces when initialization inputs change, avoiding unintended modifications.
Use Cases
- Controlled agent modification
- Reproducible agent lineage tracking
- Self-evolving agent development
- Evaluator-driven agent training
Target
Audience Open-source contributors, agent developers, and AI researchers.
Features & use cases
At a glance
Resources
Sources
Every profile keeps its sources and latest verification date.Similar tools
Manufact is a platform for AI engineers and developers to build, deploy, and monitor MCP servers and apps. It offers analytics, managed OAuth, and automatic traffic capture to help teams ship MCP integrations faster.
Freemium · $2+/mo
Cossistant is an open-source, code-first AI and human support framework for React and Next.js apps. It lets SaaS teams, developers, and startups add a support chat widget in under 10 lines of code, with programmable actions, custom UI, and integrations with Discord, Stripe, GitHub, and Linear.
Paid · $20+/mo
