Open-source evolution framework for AI agents

Let your Agent evolve.
Autonomously.

Rotifer continuously evaluates how available Genes perform in real tasks, helping your Agent adopt more reliable capability modules and get stronger over time.

Send this prompt to your Agent so it can connect Rotifer MCP and verify Genes are available.

Help me configure Rotifer MCP in my current AI IDE using @rotifer/mcp-server@latest.
Follow https://rotifer.dev/docs/guides/mcp-setup, verify that Gene search works, then tell me the next step to enhance my Agent with Genes.

Run these commands in your terminal to bootstrap a Rotifer Agent workspace and launch a preset Agent.

npx -y @rotifer/playground@latest init my-agent
cd my-agent
npx -y @rotifer/playground@latest hello --template quality-advisor

A paradigm shift

The old way

1 Humans write skills

2 Hardcode into Agent

3 Capabilities stay fixed

Static, fragile

The Rotifer way

1 Genes get verified

2 Real performance is scored

3 Reliable capabilities move first

4 Agent keeps improving

Adaptive, autonomous

Rotifer evaluates Genes on real task performance, helping Agents adopt more reliable capability modules instead of depending on a fixed toolset forever.

Your Agent gets

Not more features — better outcomes

Lower cost

The system favors more efficient Genes — fewer retries, fewer redundant calls, and lower token and compute costs

Higher quality

Inefficient or unstable Genes move down the list, while capabilities proven on real tasks move first — steadier outputs, fewer errors

Zero maintenance

Capability updates and replacements are driven by data, not manual tool management — your Agent is easier to keep current

Instant scaling

When your Agent detects a capability gap, it can adopt the right Gene on demand — from one capability to full workflows

The Agent Journey

Verify first, adopt safely, improve continuously

Identify capability gap

Agent detects a missing, inefficient, or unstable capability in its current toolkit

Discover candidates via MCP

Search the Gene ecosystem through MCP Server, filtering candidates by domain, fidelity, and real performance

Verify candidate capabilities

Compare similar Genes on task quality, safety, and cost; reliable candidates move first

Adopt and keep improving

Agent adopts stronger Genes; runtime data feeds future scoring and optimization

Agent superpowers

Everything an Agent needs to discover, verify, and adopt new capabilities

01 GENE

Atomic capability

Your Agent gets plug-and-play capability units — each Gene has clear inputs and outputs, and can be tested, replaced, and upgraded independently.

02 ARENA

Performance evaluation

Your Agent automatically chooses more reliable tools — similar Genes are compared on real task performance, reducing manual curation.

03 GENOME

Workflow composition

Simple Genes can compose into complex capabilities — your Agent grows from single skills to multi-step workflows.

04 L2

Sandbox safety

Your Agent safely tries new capabilities — static analysis → sandbox simulation → controlled trial, triple-verified before going live.

05 IR

Run anywhere

Your Agent can use the same Genes locally and in the cloud — package once, run in multiple environments.

06 MCP

One-click access

Your Agent already supports MCP? No core code changes needed — instantly tap into the Gene ecosystem.

URAA

Universal Rotifer Autonomous Architecture — five layers

L4 Collective Immunity Species Memory — network-wide threat detection and defense sharing ← Species Memory

Records security incidents, malicious gene fingerprints, and defense strategies across the entire network. Threat broadcasting, temporal decay, and consensus-verified writes form a collective immune memory.

L3 Competition & Exchange Arena selection + horizontal gene transfer across agents ← Selection & Transfer

Game-theoretic Arena where genes compete on a multiplicative fitness model. High-fitness gene metadata propagates via P2P; agents pull full genes based on capability gaps.

L2 Calibration Multi-stage gene validation — static analysis, sandbox, controlled trial ← Immune System

Emulates biological thymic selection: newly synthesized or externally acquired genes must pass static analysis, sandbox simulation, and controlled live trial before entering the main execution sequence.

L1 Synthesis WASM sandbox execution + IR compiler + resource metering ← Protein Synthesis

The protocol's ribosome — transforms raw inputs into standardized gene fragments. Rotifer IR compiles to WASM for sandboxed execution with fuel metering, memory caps, and deterministic resource accounting.

L0 Kernel Immutable trust anchor — safety floor that no layer can override ← Genetic Code

The agent's root of trust. Enforces immutable security constraints, anchors core state to a tamper-proof digest, and isolates gene permissions. The only layer that does not participate in evolution.

Visual simulation: Genes pass safety checks, get compared on real performance, and help Agents adopt more reliable capabilities.

Why L0 matters

Every agent framework lets you add tools. Only Rotifer enforces an immutable safety floor.

Agent Request

L0 Kernel

PASS — Execute

BLOCK — Violation

Framework	Safety Layer	Bypassable?
LangChain	None built-in	N/A
AutoGPT	Optional guardrails	Yes
CrewAI	Role-based limits	Yes
Manus	Prompt-level hard guardrails	Yes — operator configured
OpenAI Agents SDK	I/O + tool guardrails	Yes — manually configured
Google ADK	Callback + plugin safety	Yes — opt-in
EvoMap	Soft evolution principles	Yes — aspirational
Rotifer	L0 Kernel	No — immutable

Roadmap

From first gene to full protocol — our evolution path

v0.1 Core CLI + Genesis Genes + Arena

v0.2 IR Compiler + Live Arena + NAPI Bridge

v0.3 TS to WASM Auto-Compilation (Javy)

v0.4 Cloud Binding: Publish / Search / Install

v0.5 Reputation System + Skill Import + P2P Scaffolding

v0.5.5 L0 Kernel Gate + WASM Sandbox + Compliance Testing

v0.6 Gene Detail Pages + Creator Profiles + Cold Start

v0.6.5 Cross-Binding Proof: RotiferBinding Trait + Web3 Mock + IR Validation

v0.7 Hybrid Gene + AI Dogfooding + IDE Plugin

v0.8 Security Hardening + P2P Protocol RFC + AI Docs Assistant

v0.9 P2P Discovery Layer + Economy Design

v1.0 Full Protocol Compliance + Multi-Binding

Let your Agent evolve.Autonomously.