research
Pass
Audited by Gen Agent Trust Hub on Oct 9, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes external documentation, third-party API specifications, and repository source code. This creates a potential surface for indirect prompt injection where malicious instructions embedded in external content could attempt to influence the agent's behavior.
- Ingestion points: User-specified documentation, official API references (e.g., docs.stripe.com), local source code, and library lockfiles.
- Boundary markers: The skill instructs the agent to distinguish between source facts and inference, and to trace all material claims to direct citations.
- Capability inventory: The instructions are focused on research and synthesis; the skill does not grant capabilities for code execution, system modification, or network exfiltration beyond standard browsing.
- Sanitization: No specific technical sanitization is mentioned, though the manual verification steps serve as a functional control.
- [SAFE]: The skill provides links to official documentation and GitHub repositories for Stripe. These references target well-known and established services, which are expected in the context of integration research.
- [SAFE]: The instructions promote secure credential management by explicitly advising the use of environment-provided credentials and configured client instances rather than hardcoded keys.
Audit Metadata