The Agent Skills Directory

[COMMAND_EXECUTION]: The skill provides several examples of utilizing curl to make POST requests to the https://api.exa.ai/search endpoint. These commands are used to perform semantic searches and retrieve web data, which is the primary purpose of the skill.
[INDIRECT_PROMPT_INJECTION]: The skill acts as a gateway for external web content to enter the agent's context, creating an inherent attack surface for indirect prompt injection.
Ingestion points: The skill facilitates the retrieval of full page text, highlights, and summaries from arbitrary URLs through the Exa Search API, as detailed in the Response Fields section of SKILL.md.
Boundary markers: There are no instructions within the skill to wrap the retrieved data in protective delimiters or to explicitly warn the agent to ignore instructions embedded in the search results.
Capability inventory: The skill performs network operations via curl to fetch data from the internet.
Sanitization: The documentation does not specify any sanitization, filtering, or validation steps for the content returned by the search API before it is processed by the agent.

exa-search