The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process data from external websites, creating a surface for indirect prompt injection.\n- Ingestion points: The skill uses browser_navigate, browser_snapshot, and WebSearch tools to fetch data from competitor websites and search results as described in SKILL.md (Phases 2 and 3).\n- Boundary markers: No specific delimiters or instructions are provided in SKILL.md to advise the agent to disregard or isolate potentially malicious instructions embedded in the external content.\n- Capability inventory: The agent is granted browser interaction capabilities, including browser_click, browser_type, and browser_take_screenshot, along with reporting capabilities as defined in SKILL.md.\n- Sanitization: There are no instructions in SKILL.md for sanitizing, filtering, or validating the content retrieved from external web sources before it is processed by the agent.

competitor-scan