performing-cloud-penetration-testing-with-pacu

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 0.90). The prompt instructs entering real AWS Access Key ID and Secret Access Key into Pacu, shows commands that embed access-key identifiers and exports session data and retrieved Secrets Manager/SSM values, so secrets are likely to be handled and could be output verbatim.

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

• Third-party content exposure detected (high risk: 0.80). The skill explicitly instructs downloading and analyzing arbitrary S3 bucket data and other target-account resources (see SKILL.md Step 4 "Download S3 bucket data for analysis" and references to Pacu session downloads at ~/.pacu/sessions//downloads/ and the agent running Pacu modules), meaning it ingests untrusted, user-generated third‑party content that can influence subsequent exploitation decisions and tool actions.