postgres
Pass
Audited by Gen Agent Trust Hub on Feb 25, 2026
Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill executes the
psqlcommand-line utility inSKILL.mdto perform database operations. It depends on theDATABASE_URLenvironment variable for connection details. - [PROMPT_INJECTION]: The skill has an attack surface for indirect prompt injection (Category 8) because it processes untrusted data from database query results.
- Ingestion points: Output from
psqlquery execution inSKILL.mdenters the agent context. - Boundary markers: No delimiters or specific instructions are used in
SKILL.mdto distinguish query results from agent instructions. - Capability inventory: Ability to execute shell commands (
psql) and arbitrary SQL queries as defined inSKILL.md. - Sanitization: No sanitization or validation of database content is performed in
SKILL.mdbefore it is processed by the agent.
Audit Metadata