proto-api-generator
Pass
Audited by Gen Agent Trust Hub on Sep 21, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes external requirement documents and mock payloads to generate API designs and Protobuf files, creating a potential surface for indirect instructions to influence the agent's output.
- Ingestion points: Task intake logic in
SKILL.mddefines support for "Requirement + mock demo" and "Folder input" where untrusted external data enters the agent's context. - Boundary markers: The instructions do not define explicit boundary markers or instructions to ignore embedded prompts within the ingested requirement documents.
- Capability inventory: The skill is equipped to use the
TodoWritetool for task management and has the capability to write files to the local file system (e.g.,prd/API.mdanddesign/subdirectories). - Sanitization: There are no documented procedures for sanitizing or escaping the content derived from external requirement documents before it is used to generate the final API contracts.
- [EXTERNAL_DOWNLOADS]: Reference documentation includes instructions for the installation of software dependencies from the vendor's official GitHub repository.
- Evidence:
references/go-sphere-error-handling-reference.mddocuments the installation of theprotoc-gen-sphere-errorsplugin using the commandgo install github.com/go-sphere/protoc-gen-sphere-errors@latest.
Audit Metadata