aws-agentic-ai

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

• Third-party content exposure detected (medium risk: 0.30). The workflow reads and executes code/skills from S3 via s3 cp commands and uses the Claude Agent SDK to ingest external configuration files (SKILL.md, CLAUDE.md) from S3 (agent-configs), which allows outsider-authored text injection if an attacker can write to the monitored S3 bucket.