The Agent Skills Directory

[SAFE]: The skill is designed to improve agent performance by converting JavaScript-rendered documentation into Markdown. All identified external resources (sosumi.ai and @nshipster/sosumi) are associated with the project's author.
[EXTERNAL_DOWNLOADS]: The documentation provides instructions to install the @nshipster/sosumi package from the npm registry using standard tools like npx, bunx, and deno. This is an expected deployment method for the tool.
[PROMPT_INJECTION]: The skill exposes an indirect prompt injection surface by ingesting external web content via the fetchExternalDocumentation tool.
Ingestion points: SKILL.md (via fetchExternalDocumentation, fetchAppleDocumentation, and fetchAppleVideoTranscript tools)
Boundary markers: The instructions do not define specific delimiters or instructions for the agent to ignore embedded commands in the fetched Markdown.
Capability inventory: The skill itself does not define local file system or execution capabilities, limiting the potential impact of an injection.
Sanitization: No sanitization or content validation is mentioned.

sosumi