CWE-89 SQL Injection

Coverage: 100 rules in the SiteShadow rule registry target this CWE (registry v2.0.0). Regex 98Other-pattern 2 Also: Taint and heuristic analyzers may also detect related flows (see coverage for the authoritative list) Registry tagging shows intent, for sample-level behaviour and benchmarked gaps see known gaps.

What this means

SiteShadow flagged code that builds SQL using untrusted input without parameterization (string concatenation, interpolation, .format(), template literals, etc.).

Why it matters

• Data exposure: attackers can read sensitive rows (users, tokens, billing, PII).
• Data manipulation: attackers can update/delete data or escalate privileges.
• Auth bypass: many apps accidentally turn "query injection" into account takeover.

Safer examples

1) Use parameterized queries (Python)

cursor.execute("SELECT * FROM users WHERE email = %s", (email,))

2) Use parameterized queries (Node / pg)

await client.query("SELECT * FROM users WHERE email = $1", [email]);

3) If you must build dynamic SQL, only use allowlisted fragments

allowed = {"created_at", "email"}
order_by = order_by if order_by in allowed else "created_at"
sql = f"SELECT * FROM users ORDER BY {order_by} DESC"
cursor.execute(sql)  # only allowlisted identifiers are interpolated

How SiteShadow detects it (high level)

• Identifies query execution calls (ORM raw queries, execute, .query, .raw) combined with string building.
• Looks for user-controlled sources flowing into query strings (request params/body/query/env).
• Avoids flagging safe patterns when it can recognize parameterization APIs.

References

• CWE-89: https://cwe.mitre.org/data/definitions/89.html

---

← Back to Vulnerability Library