a2a-js-dev
Pass
Audited by Gen Agent Trust Hub on Mar 3, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill guides users to install the
@a2a-js/sdkand related dependencies (express,@grpc/grpc-js) via npm. These are vendor-owned resources and standard industry libraries required for the skill's stated purpose. - [PROMPT_INJECTION]: The skill architecture facilitates the processing of untrusted external data, creating a surface for indirect prompt injection.
- Ingestion points: Untrusted data enters the agent context through the
userMessageobject within theAgentExecutor.executemethod inSKILL.md. - Boundary markers: The code examples provided do not demonstrate the use of delimiters or 'ignore' instructions for the processed content.
- Capability inventory: The executor logic (as described in
SKILL.md) is intended to call AI models and run tools based on the incoming message content. - Sanitization: The implementation patterns shown do not include input validation or sanitization, relying on the developer to implement these safety measures.
Audit Metadata