The Agent Skills Directory

[PROMPT_INJECTION]: The skill involves processing untrusted external data, which creates a potential surface for indirect prompt injection. * Ingestion points: Workflow Step 1 in SKILL.md directs the agent to inspect product briefs, sites, repositories, analytics, waitlists, and other audience evidence. * Boundary markers: The instructions lack explicit delimiters or specific warnings to ignore instructions found within these external documents or sites. * Capability inventory: The skill requires the agent to read local files and potentially access external web content to verify links and claims. * Sanitization: There are no specific instructions for filtering or sanitizing content retrieved from external sources before it is processed by the agent.

product-launch