kaizen

Kailash Kaizen - AI Agent Framework

Kaizen is a production-ready AI agent framework built on Kailash Core SDK that provides signature-based programming and multi-agent coordination.

Features

Kaizen enables building sophisticated AI agents with:

• Signature-Based Programming: Type-safe agent interfaces with automatic validation and optimization
• BaseAgent Architecture: Production-ready agent foundation with error handling, audit trails, and cost tracking
• Multi-Agent Coordination: Supervisor-worker, agent-to-agent protocols, hierarchical structures
• Orchestration Patterns: 9 composable patterns (Ensemble, Blackboard, Router, Parallel, Sequential, Supervisor-Worker, Handoff, Consensus, Debate)
• Multimodal Processing: Vision, audio, and text processing capabilities
• Autonomy Infrastructure: 6 integrated subsystems (Hooks, Checkpoint, Interrupt, Memory, Planning, Meta-Controller)
• Distributed Coordination: AgentRegistry for 100+ agent systems with O(1) capability discovery
• Enterprise Features: Cost tracking, streaming responses, automatic optimization
• Memory System: 3-tier hierarchical storage (Hot/Warm/Cold) with DataFlow backend
• Security: RBAC, process isolation, compliance controls (SOC2, GDPR, HIPAA, PCI-DSS)
• Enterprise Agent Trust Protocol (v0.8.0): Cryptographic trust chains, TrustedAgent, secure messaging, credential rotation
• Performance Optimization (v1.0): 7 caches with 10-100x speedup (SchemaCache, EmbeddingCache, PromptCache, etc.)
• Specialist System (v1.0): Claude Code-style specialists and skills with .kaizen/ directory
• GPT-5 Support (v1.0): Automatic temperature=1.0 enforcement, 8000 max_tokens for reasoning
• Wrapper Composition System: Stackable cross-cutting wrappers (governance, monitoring, streaming) with enforced ordering

Quick Start

Basic Agent

from kaizen.core.base_agent import BaseAgent
from kaizen.signatures import Signature, InputField, OutputField
from dataclasses import dataclass

# Define agent signature (type-safe interface)
class SummarizeSignature(Signature):
    text: str = InputField(description="Text to summarize")
    summary: str = OutputField(description="Generated summary")

# Define configuration
@dataclass
class SummaryConfig:
    llm_provider: str = os.environ.get("LLM_PROVIDER", "openai")
    model: str = os.environ["LLM_MODEL"]
    temperature: float = 0.7

# Create agent with signature
class SummaryAgent(BaseAgent):
    def __init__(self, config: SummaryConfig):
        super().__init__(
            config=config,
            signature=SummarizeSignature()
        )

# Execute
agent = SummaryAgent(SummaryConfig())
result = agent.run(text="Long text here...")
print(result['summary'])