backend-code-review
Pass
Audited by Gen Agent Trust Hub on Sep 17, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [SAFE]: The skill is composed of static Markdown documents providing instructions and reference rules. It contains no executable scripts, shell commands, or configuration for remote dependencies.
- [INDIRECT_PROMPT_INJECTION]: The skill has an attack surface for indirect prompt injection because its primary function is to process and analyze untrusted code. Ingestion points: Backend source files in the
src/backend/directory and code snippets provided in chat. Boundary markers: Absent. Capability inventory: The agent's file system and terminal tools used during the review process. Sanitization: Absent. The risk is minimized by the skill's specific focus on analysis and the absence of malicious instructions within the skill itself. - [SAFE]: The skill includes comprehensive security guidelines to help the agent identify common vulnerabilities such as SQL injection, SSRF, command injection, and insecure deserialization in the code it reviews.
- [SAFE]: No patterns of obfuscation, hardcoded credentials, persistence mechanisms, or unauthorized data access were detected.
Audit Metadata