The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill's reference data in references/directory-list.md and references/submission-tracker-template.csv includes a recommendation to visit and submit product data to https://fitprofessionals.net. This domain is currently flagged as malicious and blacklisted by security scanners, posing a risk to users who follow the submission guide.
[PROMPT_INJECTION]: The skill exhibits an indirect prompt injection surface because it is designed to ingest and act upon the contents of local project files without proper isolation or validation.
Ingestion points: The skill reads data from .agents/product-marketing.md, .claude/product-marketing.md, or product-marketing-context.md in the user's project directory.
Boundary markers: The instructions lack delimiters or warnings to the agent to ignore instructions embedded within these marketing context files.
Capability inventory: The skill recommends and facilitates the use of curl for network requests to verify backlink status.
Sanitization: No validation or sanitization is performed on the ingested file content before it is processed by the agent to generate descriptions and plans.
[COMMAND_EXECUTION]: The skill instructs the agent to recommend the execution of shell commands, specifically curl -sIL [URL] | grep -i rel=, to verify the headers and dofollow status of external web pages. While standard for the task, this execution surface can be exploited if combined with malicious input from ingested files.

directory-submissions