The Agent Skills Directory

[COMMAND_EXECUTION]: The skill executes local shell scripts, including ./scripts/review-skill.sh, ./scripts/check-versions.sh, and ./scripts/install-skill.sh, to automate audit tasks and version checks.
[INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process untrusted data from external sources, such as official documentation websites and other skill files within the repository. Because it possesses the Bash, Write, and Edit tools, it is vulnerable to indirect prompt injection if the ingested content contains malicious instructions intended to manipulate the agent's behavior.
Ingestion points: Reads skill files (SKILL.md, README.md, scripts) and fetches external content via WebFetch and WebSearch from documentation sites, npm, and GitHub.
Boundary markers: The instructions do not define explicit delimiters or "ignore embedded instructions" warnings when processing external data.
Capability inventory: Full access to Bash, Write, Edit, Read, Glob, and Grep tools.
Sanitization: No specific sanitization or filtering logic is described for the content retrieved from external sources.
[EXTERNAL_DOWNLOADS]: The skill fetches data from external documentation URLs, the npm registry, and GitHub repositories to verify API patterns and package versions. These references target well-known technology services and are consistent with the skill's primary purpose.

skill-review