Skip to main content

Kinnoo agent CLI and scaffolding tools

Project description

Kinnoo

Status: Beta Python License

Kinnoo is a CLI-first platform for packaging, sharing, and installing AI agents with explicit manifest and trust controls.

Kinnoo is the lifecycle, security, and distribution harness for AI agents: define a standard manifest, package reproducible artifacts, enforce trust checks, and distribute through a registry so agents can be installed and run consistently across environments.

It is designed for AI agent developers building CLI-runnable agents who need one reliable control plane across local development and registry distribution. Kinnoo provides a standard manifest contract (kinnoo.yaml), reproducible archive workflows (.kno), trust gates for publish/install, and consistent install/run paths for end users.

What Kinnoo Is

  • Strong lifecycle harness: scaffold, run, test, inspect, pack, fetch, publish, install, and uninstall with one CLI.
  • Strong contract harness: kinnoo.yaml provides a framework-agnostic manifest with validation and predictable runtime expectations.
  • Strong trust harness: integrity verification, optional signing, strict trust gates, and inspection-first workflows before install/run.
  • Strong distribution harness: registry auth, publish/search/list/install flows, and versioned artifact distribution for teams.
  • Framework portability: build and ship agents across common framework families with consistent packaging and operator UX.

What Kinnoo Is Not

  • Not a foundation model or model-hosting service.
  • Not a replacement for framework-level orchestration (planner logic, tool routing, memory/retrieval strategy).
  • Not a UI-only chatbot builder; Kinnoo focuses on CLI-runnable agent lifecycle workflows.
  • Not a guarantee of agent quality by itself; it gives reproducibility and trust controls and provides a framework for running tests on your agents, but evaluation quality still depends on your tests and runtime design.

Installation

python3 -m venv .venv
source .venv/bin/activate
python3 -m pip install --upgrade pip
pip install kinnoo

Quick Start

kinnoo init chatgpt my-agent
kinnoo run ./my-agent "hello"
kinnoo pack ./my-agent
kinnoo login
kinnoo publish ./my-agent --pack --strict --remote

Documentation

  • docs/getting-started.md
  • docs/registry-guide.md
  • docs/kinnoo-yaml-spec.md
  • docs/cli-reference.md
  • docs/security-model.md
  • docs/supported-agents.md

Supported Frameworks

Kinnoo currently scaffolds and runs agent templates for these framework families:

  • generic (no framework)
  • chatgpt
  • gemini
  • claude-chat
  • pydantic-ai
  • langgraph
  • openai-agents
  • mcp-client
  • mcp-server
  • openclaw

See docs/supported-agents.md for capability matrix details.

Contributing

Contributions are welcome. See CONTRIBUTING.md for development and contribution workflow.

License

MIT (LICENSE).

Project details


Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distribution

kinnoo-0.8.1.tar.gz (185.1 kB view details)

Uploaded Source

Built Distribution

If you're not sure about the file name format, learn more about wheel file names.

kinnoo-0.8.1-py3-none-any.whl (203.2 kB view details)

Uploaded Python 3

File details

Details for the file kinnoo-0.8.1.tar.gz.

File metadata

  • Download URL: kinnoo-0.8.1.tar.gz
  • Upload date:
  • Size: 185.1 kB
  • Tags: Source
  • Uploaded using Trusted Publishing? No
  • Uploaded via: twine/6.2.0 CPython/3.14.3

File hashes

Hashes for kinnoo-0.8.1.tar.gz
Algorithm Hash digest
SHA256 019d0d131576d84cd3c512ceb4d8afaf6127b1b9d0569561c939fe825866663d
MD5 53d22f9152aaffa7aadc169a35b41f09
BLAKE2b-256 12f6ad9d6e2a52f36ebbe6cb854ca4c8ec302d4e8a8fe6e6b9eb4ca29439572f

See more details on using hashes here.

File details

Details for the file kinnoo-0.8.1-py3-none-any.whl.

File metadata

  • Download URL: kinnoo-0.8.1-py3-none-any.whl
  • Upload date:
  • Size: 203.2 kB
  • Tags: Python 3
  • Uploaded using Trusted Publishing? No
  • Uploaded via: twine/6.2.0 CPython/3.14.3

File hashes

Hashes for kinnoo-0.8.1-py3-none-any.whl
Algorithm Hash digest
SHA256 f4a349d9d66ee8c27a472494534f1bd639289a0ed5a1891de2db2b32a749c040
MD5 0b86ded5ddceb9c60b2b438e242be5d6
BLAKE2b-256 235af97767c53e349cd478411d8e6ffe772f936d8a267fb47ead530991fe9402

See more details on using hashes here.

Supported by

AWS Cloud computing and Security Sponsor Datadog Monitoring Depot Continuous Integration Fastly CDN Google Download Analytics Pingdom Monitoring Sentry Error logging StatusPage Status page