The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill reviews documentation files provided by the user, which could contain embedded instructions intended to manipulate the agent's behavior during the review process.
Ingestion points: Processes file or directory content identified via the $ARGUMENTS parameter using Glob and Read tools.
Boundary markers: There are no explicit instructions or delimiters used to isolate document content from the agent's core instructions.
Capability inventory: The skill utilizes Read for file access, Bash for linter execution, and WebFetch for retrieving remote style guides.
Sanitization: No sanitization or filtering of the document text is performed prior to the agent's analysis.
[COMMAND_EXECUTION]: The skill invokes the vale command-line utility to perform its primary function.
Evidence: Instructions specify running vale --output=line $ARGUMENTS and searching for the binary in local paths like /tmp/gh-aw/bin/vale.
Context: The skill's YAML frontmatter restricts Bash tool usage to vale * commands, which limits the risk of arbitrary command execution.
[EXTERNAL_DOWNLOADS]: The skill fetches guidance from external documentation pages to maintain up-to-date style rules.
Evidence: Step 1 directs the agent to retrieve style guide content from several elastic.co URLs using the WebFetch tool.
Context: All referenced URLs are hosted on the official domain of the skill's author (elastic.co), representing legitimate vendor resources.

docs-check-style