twilio-sendgrid-deliverability-advisor

Pass

Audited by Gen Agent Trust Hub on May 12, 2026

Risk Level: SAFE
Full Analysis
  • [PROMPT_INJECTION]: No attempts to override agent behavior, bypass safety guidelines, or extract system prompts were detected.
  • [DATA_EXFILTRATION]: No sensitive file paths, credential access, or network exfiltration patterns were found. The skill references well-known external domains (e.g., Google, Microsoft, Spamhaus) exclusively for informational and diagnostic purposes related to email deliverability.
  • [REMOTE_CODE_EXECUTION]: The skill does not download or execute external scripts or packages.
  • [COMMAND_EXECUTION]: There are no shell commands, subprocess calls, or dynamic execution patterns present in the analyzed files.
  • [INDIRECT_PROMPT_INJECTION]: While the skill processes user-provided symptoms and reports, it lacks exploitable capabilities (such as file writing or code execution) that would make it vulnerable to indirect injection attacks.
Audit Metadata
Risk Level
SAFE
Analyzed
May 12, 2026, 07:12 AM
Security Audit — agent-trust-hub — twilio-sendgrid-deliverability-advisor