security-expert
Pass
Audited by Gen Agent Trust Hub on Oct 8, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes data from external files (
security.jsonanddeps.json) to perform its review. If these files were to contain maliciously crafted content from an untrusted source, they could potentially influence the AI's summary output. - Ingestion points: Reads
<client-repo>/.frontend-review/report/latest/raw/security.jsonanddeps.jsoninSKILL.md. - Boundary markers: None explicitly defined in the instructions to delimit data from instructions.
- Capability inventory: Limited to reading local repository files and writing a markdown report to the local file system. No network access or command execution capabilities are requested.
- Sanitization: No specific sanitization or validation of the input JSON content is mentioned before processing.
Audit Metadata