The Agent Skills Directory

[COMMAND_EXECUTION]: The skill instructions require the agent to execute system commands (stat) to check file modification times for cache management. The commands are stat -f %m <file> for macOS and stat -c %Y <file> for Linux. The <file> path is derived from user-supplied URLs. While the skill defines a sanitization process to replace unsafe filename characters with hyphens, the use of shell commands based on external input remains a security surface.
[PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection because its core function is to ingest and process untrusted content from the web.
Ingestion points: URLs provided directly in the user prompt and URLs read from a user-specified text file on the local filesystem.
Boundary markers: The instructions do not specify the use of delimiters or clear "ignore embedded instructions" warnings when processing or storing the scraped content.
Capability inventory: The skill possesses the ability to perform network fetches, write files to the local workspace, and execute shell commands (stat).
Sanitization: The skill performs sanitization on generated filenames by replacing unsafe characters, but it does not specify any sanitization, filtering, or escaping for the actual content retrieved from the external URLs before saving it to the filesystem.

ptlam-scrapping-urls