Skip to content

README.md

Latest commit

 

History

History
93 lines (65 loc) · 2.2 KB

README.md

File metadata and controls

93 lines (65 loc) · 2.2 KB

Ada Matrix Bot

The official Matrix bot of Code Society, built with Code-Society-Lab/matrixpy.

Quick Start

  1. Create a virtualenv and install dependencies:
   python -m venv .venv
   source .venv/bin/activate
   pip install -e ".[dev]"
  1. Create a .env file at the root of the project with your credentials:
   # Username + password
   ADA_USERNAME=@yourbot:matrix.org
   ADA_PASSWORD=your_password

   # Or username + token
   ADA_USERNAME=@yourbot:matrix.org
   ADA_TOKEN=your_token
  1. Run the bot:
   ada start

By default it runs in development mode. To use a different environment:

   ada start --env production

Or point directly at a config file:

   ada start --config /path/to/config.yaml

If startup loops warnings like 'next_batch' is a required property, your token is likely invalid or expired. Regenerate ADA_TOKEN or switch to ADA_PASSWORD auth in your .env.

Configuration

Config files live in config/ and are selected by environment:

config/
  production.yaml
  development.yaml
  staging.yaml

Static values like HOMESERVER and PREFIX go in the YAML file. Secrets are never committed — they live in a .env file and are referenced as environment variable placeholders:

HOMESERVER: "https://matrix.org"
USERNAME: ${ADA_USERNAME}
PASSWORD: ${ADA_PASSWORD}
PREFIX: "!"
LOG_LEVEL: "INFO"

Extensions

Extensions are self-contained modules that add commands and event handlers to the bot. Each extension must define an extension attribute at the module level:

from matrix import Extension, Context

extension = Extension("ping")


@extension.command()
async def ping(ctx: Context) -> None:
    await ctx.reply("Pong!")

Drop the file under bot/extensions/ and it will be auto-discovered and loaded on startup — no registration needed.

Rules

  • The file must define an extension attribute, or the bot will raise a RuntimeError on startup.
  • The extension name passed to Extension() should match the module's purpose.
  • Each extension should stay focused on a single feature.