The Agent Skills Directory

[COMMAND_EXECUTION]: The skill instructs the agent to execute shell commands to manage the asset pipeline, including using node to run local character-building scripts and curl or wget for asset management.
Evidence: node scripts/build-character.mjs, node scripts/crop-head.mjs, curl -o public/assets/meme-ref.png "<image_url>".
[EXTERNAL_DOWNLOADS]: The skill automates the downloading of external assets such as company logos, meme images, and spritesheets from the web or from URLs provided in input data.
Evidence: Instructions to fetch logos from official sources and meme images from image_url fields in thread.json.
[PROMPT_INJECTION]: The skill exhibits an indirect prompt injection surface by ingesting and acting upon untrusted data from external sources and project files.
Ingestion points: The agent reads thread.json for image URLs and processes content from WebSearch results to find character headshots.
Boundary markers: No explicit boundary markers or instructions to ignore embedded commands within the processed data are provided.
Capability inventory: The skill possesses the capability to perform network requests (curl, wget), write to the file system (public/assets/), and execute local scripts (node).
Sanitization: There is no evidence of sanitization or validation of the URLs or metadata retrieved from untrusted external sources before they are used in shell commands.

game-assets