supabase-extract-db-string
Pass
Audited by Gen Agent Trust Hub on Sep 14, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted client-side source code and JavaScript bundles to identify database credentials, with the capability to write these findings directly to local files and logs.
- Ingestion points: The skill ingests external content from target application URLs and JavaScript bundles (
SKILL.md). - Boundary markers: There are no defined boundary markers or instructions to ignore embedded commands within the code being analyzed.
- Capability inventory: The skill utilizes file-writing capabilities to update
.sb-pentest-context.json,.sb-pentest-audit.log, and an evidence directory (.sb-pentest-evidence/). - Sanitization: No sanitization or validation mechanisms are mentioned to prevent malicious instructions within the target code from influencing the agent's logging actions.
Audit Metadata