Skip to main content

SwarmKit runtime: topology interpreter, LangGraph compiler, AGT-backed governance, CLI and HTTP server.

Project description

swarmkit-runtime

Python runtime for SwarmKit. Interprets topology files, compiles them into LangGraph StateGraphs, enforces governance via the GovernanceProvider abstraction (AGT-backed in v1.0), and exposes the swarmkit CLI plus a persistent HTTP server.

Layout

src/swarmkit_runtime/
├── cli/                 # Typer-based CLI: init, author, run, serve, eject
├── topology/            # Topology loader, validator, resolver (archetype + skill refs)
├── skills/              # Skill registry, category-specific semantics, composition
├── archetypes/          # Archetype registry and instantiation
├── governance/          # GovernanceProvider interface + AGTGovernanceProvider impl
├── langgraph_compiler/  # Topology → StateGraph compilation (design §14.3)
├── mcp/                 # MCP client, server lifecycle, sandbox supervision
└── audit/               # Append-only audit log adapters, skill gap log surfacing

Design references

  • §7 Architectural Principles — topology as data, eject, never lock-in
  • §8 Separation of Powers — governance/ module is the SwarmKit side; AGT is the implementation
  • §9 System Architecture — this package is component #1 of 3
  • §14 Runtime Architecture — three execution modes (one-shot, persistent, scheduled)

Entry points (design §14.2)

Command What it does
swarmkit init Launch Workspace Authoring Swarm in terminal chat mode
swarmkit author topology [name] Launch Topology Authoring variant
swarmkit author skill [name] Launch Skill Authoring Swarm
swarmkit author archetype [name] Launch Archetype Authoring variant
swarmkit run topology.yaml One-shot execution
swarmkit serve workspace/ Persistent / scheduled mode
swarmkit eject topology.yaml Export LangGraph code

Development

uv sync --package swarmkit-runtime
uv run pytest packages/runtime/tests
uv run swarmkit --help

Project details


Release history Release notifications | RSS feed

This version

1.3.0

Download files

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

Source Distribution

swarmkit_runtime-1.3.0.tar.gz (279.8 kB view details)

Uploaded Source

Built Distribution

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

swarmkit_runtime-1.3.0-py3-none-any.whl (258.4 kB view details)

Uploaded Python 3

File details

Details for the file swarmkit_runtime-1.3.0.tar.gz.

File metadata

  • Download URL: swarmkit_runtime-1.3.0.tar.gz
  • Upload date:
  • Size: 279.8 kB
  • Tags: Source
  • Uploaded using Trusted Publishing? No
  • Uploaded via: uv/0.11.17 {"installer":{"name":"uv","version":"0.11.17","subcommand":["publish"]},"python":null,"implementation":{"name":null,"version":null},"distro":{"name":"Ubuntu","version":"24.04","id":"noble","libc":null},"system":{"name":null,"release":null},"cpu":null,"openssl_version":null,"setuptools_version":null,"rustc_version":null,"ci":true}

File hashes

Hashes for swarmkit_runtime-1.3.0.tar.gz
Algorithm Hash digest
SHA256 64397d646e8882c7f65951a7e77d256b11b09ef050a398ba9dc0bee66c30fa3e
MD5 f8b87b1b1957d2a54f5e613ddf193e59
BLAKE2b-256 ed18c661c3f60d053eeecd3725efd0ff8ac4c3dfa3153aa3c8b7e8d9c75b8d8f

See more details on using hashes here.

File details

Details for the file swarmkit_runtime-1.3.0-py3-none-any.whl.

File metadata

  • Download URL: swarmkit_runtime-1.3.0-py3-none-any.whl
  • Upload date:
  • Size: 258.4 kB
  • Tags: Python 3
  • Uploaded using Trusted Publishing? No
  • Uploaded via: uv/0.11.17 {"installer":{"name":"uv","version":"0.11.17","subcommand":["publish"]},"python":null,"implementation":{"name":null,"version":null},"distro":{"name":"Ubuntu","version":"24.04","id":"noble","libc":null},"system":{"name":null,"release":null},"cpu":null,"openssl_version":null,"setuptools_version":null,"rustc_version":null,"ci":true}

File hashes

Hashes for swarmkit_runtime-1.3.0-py3-none-any.whl
Algorithm Hash digest
SHA256 d897b78fee55ec7487dc6a801af55533d296c22f4b05985e707670b21674f068
MD5 0e6c4d37be4e7da44d7a9ec17b5e3d03
BLAKE2b-256 fa5cb533ce9723e3580fc4120ebd9808b7a49bb6f706a22fc7b3cee8c2c7812a

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