twilio-sendgrid-inbound-parse
Pass
Audited by Gen Agent Trust Hub on Sep 15, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [SAFE]: No malicious patterns, obfuscation, or unauthorized data access were detected. The skill follows security best practices by providing defensive implementation guidance and correctly identifying potential risks in the documented workflow.
- [INDIRECT_PROMPT_INJECTION]: The skill describes a workflow for processing untrusted email content from external senders, which represents an indirect prompt injection attack surface.
- Ingestion points: SKILL.md documents a webhook endpoint that receives email headers, bodies, and attachments.
- Boundary markers: The documentation explicitly warns developers to isolate content as user input and avoid concatenating it directly into system prompts.
- Capability inventory: No active executable tools or dangerous capabilities are provided by the skill itself; it focuses on educational implementation snippets.
- Sanitization: The documentation recommends HTML sanitization to prevent XSS and ECDSA signature verification to ensure payload authenticity.
Audit Metadata