The Agent Skills Directory

[PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection because it ingests and processes untrusted data from external sources such as competitor websites, social media posts, and customer reviews.
Ingestion points: Data is gathered via web searches and the competitor-intel tool, including homepage headlines, value propositions, and social media content.
Boundary markers: The instructions do not specify the use of delimiters or clear separation between system instructions and untrusted data during analysis.
Capability inventory: The skill utilizes web-search and writes persistent files to the local filesystem (clients/ directory).
Sanitization: There is no evidence of content filtering or sanitization of external text before it is analyzed by the agent, which could allow maliciously crafted text on a competitor's site to influence the agent's output.

competitive-strategy-tracker