backend-db-performance
Pass
Audited by Gen Agent Trust Hub on Aug 25, 2026
Risk Level: SAFEPROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection as it ingests untrusted user input to perform database operations.
- Ingestion points: User-provided descriptions of performance issues, table names, or query fragments mentioned in the context of database optimization.
- Boundary markers: Absent; the skill does not specify the use of delimiters or instructions to ignore embedded commands within user-supplied data.
- Capability inventory: The skill has the capability to execute SQL queries (e.g., querying
pg_indexesandpg_stat_statements) and inspect the file system (vials). - Sanitization: The provided templates do not include explicit steps for sanitizing or validating user-provided database object names before they are interpolated into SQL or shell commands.
- [COMMAND_EXECUTION]: The skill instructs the agent to execute shell commands to inspect the local development environment.
- Evidence: Use of
ls -la supabase/migrations/ | grep -i "index\|optim\|perf"to check for existing database optimizations.
Audit Metadata