The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill facilitates the ingestion and processing of untrusted user content and external data sources.
Ingestion points: User-supplied text within the messages payload and external resource URLs (e.g., https://example.com/inspiration-photo.jpg) passed to the multimodal content format in SKILL.md.
Boundary markers: The prompt templates lack explicit delimiters (e.g., XML tags or triple quotes) or system instructions to ignore potentially malicious commands embedded in user-provided lyrics or image metadata.
Capability inventory: The skill utilizes the Bash tool to execute curl commands and WebFetch to interact with remote services.
Sanitization: There is no evidence of input validation or escaping for the strings interpolated into the API request payloads.

lyrics-generation