neuralcleave-sdk
Typed interfaces for building NeuralCleave plugins — without installing the full gateway.
NeuralCleave's gateway depends on FastAPI, Redis, Qdrant, and SDKs for all 14 supported channels. None of that is needed to write a plugin — only to run the gateway that loads it. This package contains just the three base classes a plugin author needs, with zero third-party dependencies.
Install
pip install neuralcleave-sdk
Writing a plugin
Every plugin package registers a Plugin subclass via a NeuralCleave.plugins
entry point:
# my_plugin/plugin.py
from neuralcleave_sdk import Plugin, PluginMetadata, Tool, ToolResult
class WeatherTool(Tool):
name = "get_weather"
description = "Get the current weather for a city."
parameters = {
"city": {"type": "str", "description": "City name", "required": True},
}
permissions = ["network"]
async def execute(self, city: str) -> ToolResult:
# ... call a weather API ...
return ToolResult(tool=self.name, output=f"Sunny in {city}")
class WeatherPlugin(Plugin):
metadata = PluginMetadata(
name="NeuralCleave-weather",
version="1.0.0",
plugin_type="tool",
description="Adds a get_weather tool.",
permissions=["network"],
)
def get_tools(self):
return [WeatherTool()]
# my_plugin/pyproject.toml
[project.entry-points."NeuralCleave.plugins"]
NeuralCleave-weather = "my_plugin.plugin:WeatherPlugin"
Once published to PyPI and installed alongside the NeuralCleave gateway,
neuralcleave hub install neuralcleave-weather discovers and loads it.
Plugin types
plugin_type |
Implement | Contributes |
|---|---|---|
tool |
get_tools() |
One or more Tool instances |
channel |
get_channel_adapter() |
A ChannelAdapter instance |
tts / stt / memory |
— | Loaded by name; see gateway docs |
generic |
on_load() / on_unload() |
Lifecycle hooks only |
Writing a channel adapter plugin
from neuralcleave_sdk import ChannelAdapter, InboundMessage
class MyChannelAdapter(ChannelAdapter):
channel_id = "my_channel"
async def connect(self) -> None: ...
async def disconnect(self) -> None: ...
async def send(self, target, text, *, reply_to=None, attachments=None):
...
async def _on_platform_event(self, raw_event: dict) -> None:
await self._dispatch(InboundMessage(
channel=self.channel_id,
sender_id=raw_event["user_id"],
sender_name=raw_event["user_name"],
text=raw_event["text"],
))
License
MIT
Release files for neuralcleave-sdk 0.1.3
For a detailed explanation of source distributions (sdists) and built distributions (wheels), please see the package formats documentation.
Source distribution (sdist)
| File | Size | Uploaded | |
|---|---|---|---|
| neuralcleave_sdk-0.1.3.tar.gz | 8.5 kB | Details |
Built distribution (wheel)
| File | Interpreter | ABI | Platform | Reset |
|---|---|---|---|---|
| neuralcleave_sdk-0.1.3-py3-none-any.whl | Python 3 | none | any | Details |
Total release size: 16.9 kB
Release files / neuralcleave_sdk-0.1.3.tar.gz
| Download URL | neuralcleave_sdk-0.1.3.tar.gz |
|---|---|
| Size | 8.5 kB |
| Tags | Source |
|
SHA-256 checksum How to use checksums |
ea4d42931e17caeca8ac957c51d8b7092329299d5af702a41674c1ac852a3a3f
|
|
BLAKE2b-256 checksum How to use checksums |
127235fd1c8fd66122efa3db570fa6da13663c079d7ec53989870f494b25b8e8
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
Yes |
| Uploaded via |
twine/6.1.0 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 Jul 23, 2026.
Transparency logRelease files / neuralcleave_sdk-0.1.3-py3-none-any.whl
| Download URL | neuralcleave_sdk-0.1.3-py3-none-any.whl |
|---|---|
| Size | 8.4 kB |
| Tags | Python 3 |
|
SHA-256 checksum How to use checksums |
3e9ca669ae8b404aee2f683916fac0bb068ae764c44a77432f5bf229b8948638
|
|
BLAKE2b-256 checksum How to use checksums |
4add3a203da043cd89a103c81bd319d0383411ddeb91eb7592f8ff86deb7e135
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
Yes |
| Uploaded via |
twine/6.1.0 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 Jul 23, 2026.
Transparency log