kaizen-agents
PACT-governed autonomous agent engines built on the Kailash Kaizen SDK.
Architecture
Layer 1: PRIMITIVES (kailash-kaizen) — deterministic, no LLM
Layer 2: ENGINES (kaizen-agents) — LLM judgment ← this package
Layer 3: ENTRYPOINTS (kz CLI, aegis) — human interface
kaizen-agents provides the LLM-driven intelligence layer on top of kailash-kaizen primitives (Signature, BaseAgent, memory, tools).
Installation
pip install kaizen-agents
This automatically installs kailash-kaizen and kailash-pact as dependencies.
Quick Start
from kaizen_agents import Agent
agent = Agent(model="claude-sonnet-4-6", budget_usd=5.0)
result = await agent.run("Analyze this codebase")
What's Included
Specialized Agents
ReAct, RAG, Tree-of-Thoughts, Chain-of-Thought, planning, vision, audio, streaming, batch processing, and more.
Multi-Agent Patterns
Debate, supervisor-worker, consensus, ensemble, pipeline, sequential, parallel, handoff, blackboard.
Journey Orchestration
Multi-pathway user journeys with LLM-based intent detection and context accumulation.
Orchestration Engine
Planner (designer, composer, decomposer), protocols (delegation, escalation, clarification), recovery (diagnoser, recomposer), monitoring.
PACT Governance
GovernedSupervisor with accountability tracking, budget enforcement, clearance checks, cascade governance.
Delegate
Autonomous TAOD engine (Think-Act-Observe-Decide) for CLI and interactive agent applications.
License
Apache 2.0 — Terrene Foundation
1 maintainer
Metadata
1 maintainer
Release files for kaizen-agents 0.13.0
For a detailed explanation of source distributions (sdists) and built distributions (wheels), please see the package formats documentation.
Built distribution (wheel)
| File | Interpreter | ABI | Platform | Reset |
|---|---|---|---|---|
| kaizen_agents-0.13.0-py3-none-any.whl | Python 3 | none | any | Details |
Release files / kaizen_agents-0.13.0-py3-none-any.whl
| Download URL | kaizen_agents-0.13.0-py3-none-any.whl |
|---|---|
| Size | 699.7 kB |
| Tags | Python 3 |
|
SHA-256 checksum How to use checksums |
dff65e1c11afcaccf64d790b74a7adf922b361876414d621dfb04e8f1f777baa
|
|
BLAKE2b-256 checksum How to use checksums |
c4138444dd96a7f1876f7b276312fc142890749032bcffdd68cb5a230a98fb58
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
Yes |
| Uploaded via |
twine/7.0.0 CPython/3.13.14
|
Provenance
Provenance describes where a file came from. On PyPI, provenance is shared via attestations, which provide a verifiable record of the build or publishing details. View details, limitations and caveats.
PyPI Publish Attestation
PyPI verified that this artifact, at this checksum, originated from the publisher listed below.
Signed by GitHub Actions, verified by PyPI on Aug 17, 2026.
Transparency log
