The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill processes data from external web search results provided by the XCrawl API.
Ingestion points: The https://run.xcrawl.com/v1/search API response body, which contains untrusted content from various websites.
Boundary markers: The instructions lack explicit delimiters or warnings to treat the search result content as untrusted data, although it advises returning the raw response without synthesis.
Capability inventory: The skill possesses significant capabilities including Bash (curl/node), Read, Write, and Edit tools, which could potentially be misused if the agent obeys instructions embedded in the ingested search results.
Sanitization: There is no evidence of sanitization or filtering of the external content before it is processed by the agent.
[COMMAND_EXECUTION]: The skill utilizes node and curl via the Bash tool to interact with the API and read local configuration.
Evidence: The skill uses node -e to read and parse the XCRAWL_API_KEY from ~/.xcrawl/config.json and uses curl to transmit data to the vendor's API endpoint at https://run.xcrawl.com/v1/search.

xcrawl-search