The Agent Skills Directory

[PROMPT_INJECTION]: The skill is subject to indirect prompt injection risks because it ingests and processes data from external websites. \n- Ingestion points: Untrusted data enters the agent context through the mcp__playwright__browser_navigate and DOM inspection steps described in SKILL.md. \n- Boundary markers: None; the skill lacks delimiters or specific instructions to the agent to ignore commands potentially embedded in the scanned content. \n- Capability inventory: The skill employs mcp__playwright tools which allow for browser navigation and network request logging. \n- Sanitization: None; website content is analyzed as-is without validation or escaping techniques to mitigate prompt injection.

mixed-content-scan