The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill processes documentation content fetched from external, potentially untrusted sources via curl and WebFetch. This creates a vulnerability surface where malicious instructions could be embedded in the documentation files.
Ingestion point: curl fetches content from target URLs in Step 1 and Step 4.
Boundary markers: The skill includes checkpoints for availability but does not specify explicit prompt delimiters or 'ignore' instructions for the ingested content.
Capability inventory: curl (network/write), grep, cp, head (file system), skill-seekers (external tool).
Sanitization: Content validation is limited to checking for error pages and file size rather than content safety.
[EXTERNAL_DOWNLOADS]: The skill uses curl to fetch documentation from remote URLs. This behavior is central to its stated purpose of ingesting LLM-optimized documentation from the web.
[COMMAND_EXECUTION]: The skill utilizes shell commands including curl, grep, head, and cp to interact with the file system and perform network requests as part of its documented workflow.

llms-txt-support