The Agent Skills Directory

[SAFE]: The skill performs its stated purpose using trusted resources from the Elastic organization. It fetches style guidance from official domains and uses a restricted toolset.\n- [COMMAND_EXECUTION]: The skill executes the vale linter via Bash using user-provided arguments. Although the allowed-tools configuration limits commands to the vale prefix, the interpolation of $ARGUMENTS into a shell command represents a potential command injection surface that relies on platform-level enforcement for safety.\n- [PROMPT_INJECTION]: The skill ingests and analyzes user-provided Markdown files. This represents an indirect prompt injection surface as malicious instructions could be embedded in the documentation being reviewed. The agent is not provided with specific delimiters or instructions to ignore embedded commands during the review process.\n
Ingestion points: Markdown files read from the path provided in $ARGUMENTS.\n
Boundary markers: No delimiters or ignore-instructions are used when processing file content.\n
Capability inventory: Bash (restricted), Read, Grep, Glob, WebFetch, and CallMcpTool.\n
Sanitization: No content sanitization is performed.

docs-check-style