The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill instructs the agent to fetch required-reason API documentation and other guidelines from an unofficial third-party domain, which is not associated with the official platform or recognized trusted vendors.
Evidence in SKILL.md: https://sosumi.ai/documentation/bundleresources/describing-use-of-required-reason-api
Evidence in references/privacy-manifest.md: Multiple links to sosumi.ai for API categories and reason codes (e.g., https://sosumi.ai/documentation/bundleresources/app-privacy-configuration/nsprivacyaccessedapitypes/nsprivacyaccessedapitype).
[INDIRECT_PROMPT_INJECTION]: The skill exhibits an attack surface for indirect prompt injection because it is designed to ingest and act upon content from external, third-party sites that the skill author does not control.
Ingestion points: SKILL.md and references/privacy-manifest.md instruct the agent to "fetch the current App Review Guidelines, Upcoming Requirements, screenshot specifications, and required-reason API documentation" from external URLs.
Boundary markers: There are no explicit instructions or delimiters provided to the agent to treat the fetched content as untrusted or to ignore any potential instructions embedded within those documents.
Capability inventory: The agent is instructed to perform a comprehensive audit, separate blockers from cleanup, and validate build archives based on the information retrieved from these external sources.
Sanitization: The skill does not mention any validation, filtering, or sanitization processes for the content retrieved from the external documentation sites.

app-store-review