maltego-transforms

The Python SDK for building Maltego transform servers.

---

A transform takes one or more Maltego entities as input — an IP address, a domain, a person, a document, or your own custom object — and returns related entities as output. Transforms are how Maltego expands a graph, and maltego-transforms is the fastest way to write and run your own.

Install

pip install maltego-transforms

Requires Python >=3.10,<3.15.

Your first transform

from maltego.server import (
    MaltegoContext,
    MaltegoEntity,
    MaltegoServerSettings,
    register_transform,
    run_server,
    setup,
)


@register_transform(
    display_name="Hello World [My Transforms]",
    transform_set="My Transforms",
)
async def hello_world(
    input_entity: MaltegoEntity,
    context: MaltegoContext,
) -> MaltegoEntity:
    context.log.inform(f"Hello from {context.remote_ip}")
    return MaltegoEntity["maltego.Phrase"](f"Hello, {input_entity.value}")  # type: ignore


if __name__ == "__main__":
    settings = MaltegoServerSettings(
        server_name="My Transform Server",
        ns="my.transforms",
        author="i@example.com",
    )
    setup(settings)
    run_server(settings=settings, ssl=False)  # plain HTTP for local development

Save this as project.py and run it:

python project.py

This starts a local HTTP server on 127.0.0.1:3000, serving a seed URL at /seed. Add that URL to Maltego to discover and install your transforms. SSL is on by default for anything beyond local dev — see HTTPS, Certificates, and Browser Trust, which the Graph Browser client requires.

Prefer a ready-made project over a blank file?

maltego-transforms start my_project
cd my_project
python -m venv .venv && source .venv/bin/activate
pip install -r requirements.txt
python project.py

This scaffolds a runnable project with example transforms covering the most common patterns.

Runbooks

Repository runbooks live in runbooks/ and capture repeatable workflows for contributors and coding agents, including writing PRPs for non-trivial changes and choosing the right test command.

Features

• Decorator-based registration — @register_transform turns any function into a discoverable transform
• Typed input and output — plain Python type hints; the SDK infers and validates entity types for you
• Async and streaming — write async def transforms, or use AsyncGenerator to stream results as they're found
• OAuth — built-in support for OAuth-authenticated transforms
• Transform settings — expose configurable options (API keys, limits, toggles) that users fill in from the Maltego client
• Input constraints — control which entities a transform is even offered for, before it runs
• Middlewares — hook into the request/response lifecycle for logging, auth, or shared setup

Standard entities

Want typed entity classes instead of raw dictionaries? maltego-transforms-std-entities ships ready-made classes like Person, DNSName, and IPv4Address:

pip install maltego-transforms-std-entities

from maltego.entities import DNSName, IPv4Address
from maltego.server import register_transform

@register_transform
async def dns_to_ip(input_entity: DNSName) -> IPv4Address:
    return IPv4Address("1.1.1.1")

Building with an AI coding agent

Scaffold a project with agent skills for authoring and testing transforms:

maltego-transforms start my-project --with-skills

This drops a skill index into .agents/skills/ that points your coding agent at the right focused skill for the task at hand — building a new transform, migrating an existing one, or debugging discovery. See Using AI Agent Skills with the SDK for details.

Documentation

Full docs — configuration, authentication, entity modeling, pagination, and the complete API reference — start at the Maltego Transforms SDK overview.

Support

Questions or issues? Reach out through the Maltego Support Portal or your Maltego account contact.

Contributing

Contributions are welcome — see CONTRIBUTING.md for guidelines.

License

maltego-transforms is licensed under the MIT License.