The Agent Skills Directory

[SAFE]: The skill implements a design review workflow based on official Google API Improvement Proposals (AIPs). It correctly directs the agent to authoritative documentation at google.aip.dev and includes standard security guidance, such as avoiding the return of sensitive fields as specified in AIP-147.
[INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and analyze external protobuf and OpenAPI definitions. 1. Ingestion points: Defined in the frontmatter of SKILL.md for .proto and openapi yaml/json files. 2. Boundary markers: The instructions do not define specific delimiters for separating the design instructions from the contents of the analyzed files. 3. Capability inventory: The skill possesses no capabilities for executing commands, performing network operations to untrusted domains, or writing to sensitive system files. 4. Sanitization: Not applicable as the skill's primary function is static design analysis and reporting.

aip