The Agent Skills Directory

[PROMPT_INJECTION]: The skill is vulnerable to indirect prompt injection because it reads and processes untrusted lyrics data from the filesystem without established security boundaries.
Ingestion points: Lyrics text is ingested via the extract_section tool in SKILL.md from files located at the path provided in the skill's arguments.
Boundary markers: The instructions do not employ delimiters or provide specific instructions for the agent to ignore commands or instructions found within the lyrics content.
Capability inventory: The skill has access to the Read, Glob, and Grep tools, as well as the bitwize-music-mcp toolset, which can be used to navigate and read the local filesystem.
Sanitization: There is no evidence of sanitization, filtering, or validation performed on the retrieved lyrics text before it is presented to the language model for analysis.

explicit-checker