aws-cloudwatch

Warn

Audited by Snyk on Jul 27, 2026

Risk Level: MEDIUM
Full Analysis

MEDIUM W011: Third-party content exposure detected (indirect prompt injection risk).

  • Third-party content exposure detected (medium risk: 0.65). SKILL.md:13-47 shows the workflow running CloudWatch Logs Insights where the user-supplied --query-string and resulting log @message/fields are ingested from AWS into the agent-side execution (and thus into the LLM-visible context when results are returned); AWS log contents are outsider-authored text (not chosen by the operating user as their own input).

Issues (1)

W011
MEDIUM

Third-party content exposure detected (indirect prompt injection risk).

Audit Metadata
Risk Level
MEDIUM
Analyzed
Jul 27, 2026, 03:29 AM
Issues
1
Security Audit — snyk — aws-cloudwatch