spring-boot-in-action
Pass
Audited by Gen Agent Trust Hub on May 16, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill's components, including the instructional markdown and the pre-review Python script, are transparent and aligned with the stated educational and development goals. No obfuscation or remote access patterns were identified.
- [DATA_EXPOSURE_AND_EXFILTRATION]: Placeholder credentials (e.g., 'secret123' in
examples/before.mdand 'S3cr3tP@ss' inevals/evals.json) are used as examples of insecure practices that the skill is designed to flag and correct. They do not constitute a leak of real credentials. - [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and analyze user-provided code, creating a surface for indirect prompt injection.
- Ingestion points: User-supplied Java and property files ingested during the review process (SKILL.md).
- Boundary markers: None. The skill does not define specific markers to delimit untrusted code from instructions.
- Capability inventory: The skill can read local files via the provided
scripts/review.pyand agent file tools. - Sanitization: None. Inputs are processed directly via regex or NLP without sanitization.
Audit Metadata