The Agent Skills Directory

[COMMAND_EXECUTION]: The skill executes shell commands to interact with the x-cmd CLI to fetch, search, and list RFC documents (e.g., x rfc ls, x rfc 791).- [REMOTE_CODE_EXECUTION]: The installation guide in data/install.md suggests a method for downloading and executing a script directly from the vendor's domain (https://get.x-cmd.com | sh). This represents remote code execution from a vendor-controlled source.- [EXTERNAL_DOWNLOADS]: The skill facilitates the download of binaries and scripts from get.x-cmd.com, github.com/x-cmd/release, and conda.prefix.dev. It also fetches RFC document content from external repositories.- [PROMPT_INJECTION]: The skill has an indirect prompt injection surface as it retrieves and processes external RFC documents which are untrusted data sources.
Ingestion points: RFC document text is ingested via the x rfc <number> and x rfc : <keyword> commands.
Boundary markers: There are no explicit boundary markers or instructions to the agent to ignore instructions embedded within the retrieved RFC content.
Capability inventory: The skill possesses the ability to execute various subcommands within the x-cmd environment.
Sanitization: No sanitization of the retrieved document text is documented before it is processed by the agent.

x-rfc