The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill reference files (directory-list.md and submission-tracker-template.csv) include a link to https://fitprofessionals.net, which has been flagged as malicious and blacklisted by security scanners.
[COMMAND_EXECUTION]: Instructions in SKILL.md suggest that the user or agent verify backlinks using the command curl -sIL https://directory.com/your-listing | grep -i rel=. Providing instructions to execute shell commands against external URLs, particularly when the resource list contains known malicious sites, presents a risk of interacting with harmful remote endpoints.
[PROMPT_INJECTION]: The skill exhibits an indirect prompt injection surface by instructing the agent to read and adopt instructions from untrusted local files such as .agents/product-marketing.md.
Ingestion points: .agents/product-marketing.md, .claude/product-marketing.md, and product-marketing-context.md.
Boundary markers: Absent; the skill does not specify any delimiters or warnings to ignore instructions embedded in the ingested context.
Capability inventory: File system access (reading context) and shell command execution (curl).
Sanitization: None specified for the content read from external marketing files.

directory-submissions