The Agent Skills Directory

[COMMAND_EXECUTION]: The skill utilizes multiple Deno scripts located in the scripts/ directory to perform game logic and calculations.
These scripts are designed to be executed by the agent to manage entity states and world history.
The scripts use Deno's permission system (specifically --allow-read and --allow-write) to control access to the local environment.
[INDIRECT_PROMPT_INJECTION]: The skill contains a vulnerability surface where processed data could influence agent behavior during the narrative proposal generation phase.
Ingestion points: External strings such as entity names, event descriptions, and vulnerability details enter the system via command-line arguments to the Deno scripts.
Boundary markers: The skill lacks explicit boundary markers or instructions for the agent to ignore embedded commands within the ingested data before it is processed by the model for proposal synthesis.
Capability inventory: The scripts/propose-shift.ts tool possesses the capability to write generated proposals to the local file system.
Sanitization: Ingested inputs are interpolated directly into markdown strings without formal validation or escaping mechanisms.

world-fates