The Agent Skills Directory

[SAFE]: The skill instructions are purely informational and focus on marketing strategies and compliance. There are no executable scripts, remote downloads, or network operations that pose a security risk.
[PROMPT_INJECTION]: No attempts to override system prompts or safety filters were found. The skill reinforces safety by instructing the agent to adhere to FTC guidelines and reject unethical marketing requests, such as undisclosed sponsorships or deceptive comment seeding.
[DATA_EXFILTRATION]: No patterns of sensitive data access or external transmission were identified. The skill does not request access to credentials or environment variables.
[INDIRECT_PROMPT_INJECTION]: The skill involves fetching external social media content for vetting creators via tools like social-fetch. While this introduces a surface for untrusted data, the instructions focus on analytical vetting and compliance verification rather than automated execution. The presence of clear compliance guidelines (FTC disclosure) and the requirement for human-in-the-loop review of creative briefs mitigate risks associated with untrusted data ingestion.

influencer-marketing