The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill references the Vale prose linter (vale.sh) and configuration files within the Grafana 'writers-toolkit' repository on GitHub. These are standard resources for documentation workflows and are hosted on official or well-known domains.
[PROMPT_INJECTION]: The skill processes documentation content which constitutes an indirect prompt injection surface.
Ingestion points: Document files are read via the Read tool as part of the review steps defined in SKILL.md.
Boundary markers: The instructions do not define explicit delimiters or instructions to prevent the agent from following commands potentially embedded in the reviewed documentation.
Capability inventory: The agent uses Bash, Read, and Grep to perform analysis and execute development tools.
Sanitization: There is no mentioned process for sanitizing documentation content before it is read by the agent.
[COMMAND_EXECUTION]: The skill executes the vale command-line tool via the Bash tool to perform automated linting on documentation files based on provided paths.

docs-review