The Agent Skills Directory

[PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection as it is designed to retrieve and process content from external web pages.
Ingestion points: Content is ingested from arbitrary URLs via WebFetch, alma browser, and curl commands.
Boundary markers: There are no explicit instructions or delimiters defined to help the agent distinguish between its system instructions and potentially malicious instructions embedded in the retrieved web content.
Capability inventory: The skill possesses Bash and WebFetch capabilities, allowing it to perform network operations and execute system commands based on instructions it may receive.
Sanitization: The skill description does not include mechanisms for sanitizing or filtering the fetched content before it is processed by the agent.
[COMMAND_EXECUTION]: The skill utilizes the Bash tool to execute commands for networking and data processing.
It provides examples of using curl to fetch JSON APIs, download files to /tmp/, and check HTTP status codes.
It also utilizes jq for processing JSON data within the shell environment.

web-fetch