The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill references a blacklisted malicious domain, https://fitprofessionals.net, within its directory catalog (references/directory-list.md) and submission tracker (references/submission-tracker-template.csv). Automated scans have confirmed this URL belongs to a blacklist.
[COMMAND_EXECUTION]: The instructions advise using shell commands such as curl -sIL to verify the status of directory listings. This encourages the agent or user to perform network operations against untrusted and potentially malicious external targets.
[PROMPT_INJECTION]: The skill establishes a workflow where an AI agent interacts with and processes data from hundreds of external, third-party directories, creating a significant attack surface for indirect prompt injection.
Ingestion points: The agent is directed to visit and verify external URLs listed in references/directory-list.md and references/submission-tracker-template.csv.
Boundary markers: The skill does not provide instructions to the agent to treat external site content as untrusted or to use delimiters to isolate external data from the system prompt.
Capability inventory: The skill utilizes network capabilities via the curl command to interact with web servers.
Sanitization: There is no evidence of validation, filtering, or sanitization of the content retrieved from external sites before it is processed by the agent.

directory-submissions