Connection Pool Tuner
Pass
Audited by Gen Agent Trust Hub on Sep 5, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill instructs the agent to ingest and analyze external evidence from database activity queries (
pg_stat_activity), which could contain malicious instructions embedded by an attacker (e.g., in query strings or application metadata) designed to influence the agent's diagnostic output or configuration actions. - Ingestion points: Evidence is captured from
pg_stat_activityresults and existing application/proxy configurations in Step 1. - Boundary markers: The instructions do not specify the use of delimiters or 'ignore embedded instructions' warnings when processing this captured evidence.
- Capability inventory: The skill's workflow involves modifying application pool settings (e.g., HikariCP), PgBouncer transaction modes, and database parameters like
statement_timeoutandidle_in_transaction_session_timeout. - Sanitization: There are no mentioned procedures for sanitizing or validating the content of the captured database activity before analysis.
Audit Metadata