The Agent Skills Directory

[COMMAND_EXECUTION]: Executes a local validation utility located at ~/Development/brand-kit/dist/src/cli.js to process brand.yaml content. It also performs shell-based image processing using npx and a for loop.
[EXTERNAL_DOWNLOADS]: Uses npx to fetch and execute svgo and sharp-cli. These are well-known image optimization and manipulation tools from the public npm registry.
[PROMPT_INJECTION]: The skill interpolates external data into LLM instructions, creating an attack surface for both direct and indirect injection.
Ingestion points: Data is ingested from the brand.yaml file and the user-provided --prompt argument.
Boundary markers: The prompt template lacks explicit delimiters or instructions for the LLM to ignore potentially malicious commands embedded in the brand identity data.
Capability inventory: The skill has the capability to execute local scripts via node, download/execute packages via npx, and write files to the local directory (e.g., assets/ and brand.yaml).
Sanitization: There is no evidence of sanitization or escaping of the brand name, colors, or personality traits before they are interpolated into the generation prompt.

brand-logo