SQL query built from unescaped input

CriticalAV-RUNTIME-009Runtime safetyCore pack

What it finds

A SQL statement is assembled by string concatenation, formatting, or interpolation instead of bound parameters.

Why it matters

If any part of that string comes from a user, a document, or the model itself, it can rewrite the query — reading, changing, or deleting data the agent was never meant to touch.

What an attacker could do: Anyone who can influence the input (including through prompt injection) could read or modify any table this database connection can reach.

How to fix it

Use parameterized queries / prepared statements for every value, and give the agent a database role limited to the exact tables and operations it needs.

Pro shows the corrected code for your exact file, step by step, and re-scans to confirm the fix. See plans

Check details

Finding code: SQL_INJECTION_RISK
How it detects: Looks for a specific risky pattern in the code or config.
Confidence: Likely: strong signal, but review it in context.
Scans: Any text-based agent source or configuration (JS/TS/Python/JSON/YAML)
OWASP LLM Top 10: LLM02 - Insecure Output Handling, LLM08 - Excessive Agency
NIST AI RMF: MS-2.3, MG-2.2
SOC 2: CC6.1, CC7.1

Check your agent for this

Paste your agent code or MCP config. The free scan runs all 124 checks in about 10 seconds — no account needed.

Scan an agent free Run it in CI or your editor