acc-docker-multistage-knowledge

Pass

Audited by Gen Agent Trust Hub on Feb 16, 2026

Risk Level: LOW
Full Analysis
  • Prompt Injection (SAFE): No instructions found that attempt to override agent behavior or bypass safety filters. The content is strictly documentation and code snippets related to Docker.
  • Data Exposure & Exfiltration (SAFE): No hardcoded credentials, sensitive file paths (like ~/.ssh), or unauthorized network operations were detected.
  • Obfuscation (SAFE): No Base64, zero-width characters, or other encoding techniques intended to hide malicious content were found.
  • Unverifiable Dependencies & Remote Code Execution (SAFE): The Dockerfile snippets use official, well-known images (composer, php-alpine). Package installations (apk, composer, pecl) are standard for the context and do not pull from untrusted or suspicious sources.
  • Privilege Escalation (SAFE): The skill follows security best practices by recommending a non-root user (USER www-data) for the production stage. No use of sudo or insecure chmod patterns found.
  • Persistence Mechanisms (SAFE): No commands found that attempt to modify system startup, cron jobs, or shell profiles.
  • Metadata Poisoning (SAFE): Metadata fields are descriptive and accurately reflect the content of the skill.
  • Indirect Prompt Injection (SAFE): This skill provides static educational content and does not ingest untrusted external data at runtime, posing no risk of indirect injection.
  • Time-Delayed / Conditional Attacks (SAFE): No logic found that triggers behavior based on time, date, or specific environmental conditions.
  • Dynamic Execution (SAFE): No usage of eval(), exec(), or runtime code generation from untrusted sources was detected.
Audit Metadata
Risk Level
LOW
Analyzed
Feb 16, 2026, 09:01 AM