The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill includes a reference to a blacklisted malicious URL in its directory catalog, which could lead the agent or user to interact with a high-risk domain.
Evidence: The domain https://fitprofessionals.net is listed as a target in references/directory-list.md and references/submission-tracker-template.csv.
[COMMAND_EXECUTION]: The instructions direct the agent to execute shell commands to verify the presence and type of backlinks on external websites.
Evidence: The command curl -sIL https://directory.com/your-listing | grep -i rel= is provided in SKILL.md as part of the submission verification workflow.
[PROMPT_INJECTION]: The skill creates a surface for indirect prompt injection by instructing the agent to read and follow instructions from external, untrusted marketing context files without specifying boundary markers or sanitization routines.
Ingestion points: Instructions in SKILL.md to read .agents/product-marketing.md, .claude/product-marketing.md, or product-marketing-context.md before proceeding.
Boundary markers: Absent; there are no instructions to the agent to treat the content of these files as data rather than instructions.
Capability inventory: The skill empowers the agent to manage a multi-step distribution workflow involving network requests (link verification) and data processing across multiple reference files.
Sanitization: Absent; no filtering or validation is performed on the ingested context files.

directory-submissions