analyzing-timescaledb
Warn
Audited by Gen Agent Trust Hub on Apr 12, 2026
Risk Level: MEDIUMCOMMAND_EXECUTIONDATA_EXFILTRATIONPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill provides shell script templates that execute the
psqlutility with direct variable interpolation. Specifically, thetsdb_queryhelper function and associated scripts use variables like$HYPERTABLEwithin SQL strings (e.g.,WHERE hypertable_name = '$HYPERTABLE'). This creates a risk of SQL injection if an attacker provides a hypertable name containing malicious SQL commands (e.g.,mytable'; DROP TABLE users; --). - [DATA_EXFILTRATION]: The skill accesses sensitive database configuration and data samples using the
$TIMESCALE_URIenvironment variable, which typically contains connection credentials. The instructions mandate multi-phase discovery that reveals internal schema details, compression settings, and sample records, which are then outputted to the agent's context. - [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection (Category 8). It ingests untrusted data from the database (via
psqloutput) and processes it without sanitization or explicit boundary markers. If a database contains malicious content in hypertable names or column metadata, it could influence the agent's subsequent reasoning and actions.
Audit Metadata