swarmkit 0.1.33

Pythonic wrapper around the TypeScript SwarmKit SDK for multi-agent orchestration in E2B sandboxes

SwarmKit Python SDK

SwarmKit lets you run and orchestrate terminal-based AI agents in secure sandboxes with built-in observability.

Get Started

1. Get your SwarmKit API key at dashboard.swarmlink.ai (new users: request access first)

2. Get an E2B API key at e2b.dev for sandbox execution

3. Install the SDK:

   pip install swarmkit
   

   Note: Requires Node.js 18+ (the SDK uses a lightweight Node.js bridge).

4. Check out the official documentation and cookbooks to start shipping with SwarmKit!

Streaming Events

You can subscribe to streaming events:

• stdout: Raw NDJSON output from the agent/CLI (may arrive in chunks).
• stderr: Process stderr (may arrive in chunks).
• content: Parsed ACP-style events (messages, tool calls, plan updates).

Large Outputs

The Node bridge enforces per-event size limits for responsiveness.

• stdout/stderr are chunked when very large; your callback receives each chunk normally.
• content events are lossless: if a payload is too large to stream safely, you receive a truncated preview plus a ref to a local JSON file containing the full payload.

Example:

def on_content(evt):
    if evt.get("truncated") and evt.get("ref"):
        full_update = open(evt["ref"]["path"], "r", encoding="utf-8").read()
        # full_update is the original ACP update JSON

swarmkit.on("content", on_content)

Reporting Bugs

We welcome your feedback. File a GitHub issue to report bugs or request features.

Connect on Discord

Join the SwarmKit Developers Discord to connect with other developers using SwarmKit. Get help, share feedback, and discuss your projects with the community.

License

See LICENSE for details.