A Telegram bot for simulating and role-playing global risk scenarios. Players can take on different roles and interact with scenarios through actions and information sharing. You can read more about the motivation behind it here: Scaling Wargaming for Global Catastrophic Risks with AI.
- Clone this repository
- Install dependencies: with
bun install - Create a
.envfile with the following tokens:
TELEGRAM_BOT_TOKEN=your_key
ANTHROPIC_API_KEY=your_key
You can get an Anthropic API key from console.anthropic.com, which provides access to Claude Sonnet 3.7, the model used by this application. You can get a Telegram bot token by following instructions here (messaging the @BotFather account on Telegram.)
- Start the bot:
bun src/grim.ts. Optionally, you can provide a scenario file,bun src/grim.ts --scenario-file file_name. There is an example scenario file calledscenario.example.txtif you want to get up and running quickly. - Set your role with
/role <Your Name> - <Your Role> - Start the scenario with
/scenario <scenario_description>. You don't need to providescenario_descriptionif you used ascenario_file - Once the scenario is started, you can use these commands:
/info- Queue an information request about the current situation/feed- Queue information to incorporate into the world state/action- Queue an action you want to take in the world/process- Process all queued actions/remove- Remove an item from the action queue/rollback- Roll back the scenario to a previous checkpoint
We're happy to do some hand-holding to onboard a contributer who may want to contribute consistently! Feel free to ask for clarification in any of the outstanding issues or request a call. If you request a call, quickly describe your background or provide a link to your work. Otherwise, feel free to fork the repo, make pull requests, or make issues suggesting improvements.
Distributed under the GPL. If this is a hurdle for you, let us know.