The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill maintains a comprehensive list of third-party directories in references/directory-list.md and references/submission-tracker-template.csv. One specific entry, https://fitprofessionals.net, is flagged as a malicious URL by automated scanners. Users following the skill's instructions to submit data to this directory may be exposed to phishing or other web-based threats.
[PROMPT_INJECTION]: The skill instructions in SKILL.md mandate reading external context from files such as .agents/product-marketing.md or .claude/product-marketing.md. This creates an indirect prompt injection surface where an attacker could place malicious instructions in these marketing context files to manipulate the agent's output or actions during the directory submission workflow.
[COMMAND_EXECUTION]: The workflow in SKILL.md includes a step for users to verify backlinks using the shell command curl -sIL [URL] | grep -i rel=. While the intended purpose is a benign header check, it encourages the execution of commands against arbitrary external URLs, which is a common pattern in exploitation chains.

directory-submissions