Prompt-injection style instruction pattern detected.
Warn
- Code
- suspicious.prompt_injection_instructions
- Location
- skills/postgres-aiops/references/agent-guardrails.md:31
- Evidence
Copy this into your agent's system prompt:
Security audit
Postgres AIops
Security checks for vulnerabilities and agentic risk
Overview
This plugin clearly describes a PostgreSQL operations tool with disclosed database read/write powers, credential handling, audit logging, and no hidden or unrelated behavior in the inspected artifacts.
Install this only for PostgreSQL environments where you want an agent to inspect and potentially change database state. Start with a read-only or least-privilege PostgreSQL role, review dry-run output before maintenance actions, and treat irreversible operations such as terminate/cancel, VACUUM FULL, REINDEX, and query-stat resets as production-impacting even though they are audited.
SkillSpector was not run because this plugin release contains no bundled skills.
Static analysis
Detected: suspicious.prompt_injection_instructions