The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill facilitates the ingestion of external data from the Google Serper API. This content is untrusted and could potentially contain malicious instructions designed to influence the agent's behavior.
Ingestion points: Search results returned by the serper_search tool are processed and summarized by the AI agent.
Boundary markers: The instructions do not define specific delimiters or security warnings to be used when presenting search results to the LLM.
Capability inventory: The agent possesses network search capabilities and is documented as having access to related tools like web_fetch, which can be used to retrieve full page content.
Sanitization: There is no evidence of explicit sanitization or filtering of the search snippets or metadata before they are incorporated into the conversation context.

serper-search