The Agent Skills Directory

[COMMAND_EXECUTION]: The skill primarily operates by executing a custom utility named sw-map through the Bash tool. This utility handles map generation, location validation, and resource listing within the skill's environment.
[EXTERNAL_DOWNLOADS]: The skill uses the fal-ai/flux-2 model via the fal.ai service to generate images. This is a well-known image generation service and its use is a documented feature of the skill.
[PROMPT_INJECTION]: The skill creates a surface for indirect prompt injection by accepting user-provided input—such as location names, landscape features, and combat scenes—which are subsequently used to construct prompts for an LLM (Claude Haiku 3.5).
Ingestion points: Arguments passed to the sw-map generate command, specifically the <name>, --features, and --scene flags.
Boundary markers: No explicit delimiters or instructions to ignore embedded commands are visible in the provided skill text.
Capability inventory: The skill possesses the capability to execute shell commands (sw-map) and perform file system writes (saving JSON and PNG files to the data/maps/ directory).
Sanitization: The skill performs basic validation of location names against a local geography.json file using fuzzy matching, which provides a layer of data integrity.

map-generator