Raw vector
CVSS:4.0/AV:N/AC:L/AT:N/PR:L/UI:P/VC:H/VI:H/VA:H/SC:H/SI:H/SA:H/E:X/CR:X/IR:X/AR:X/MAV:X/MAC:X/MAT:X/MPR:X/MUI:X/MVC:X/MVI:X/MVA:X/MSC:X/MSI:X/MSA:X/S:X/AU:X/R:X/V:X/RE:X/U:XSummary
CVE-2026-17351 is a critical-severity SQL Injection (CWE-89) vulnerability in Pgadmin Pgadmin 4. Its CVSS base score is 9.4 (Critical).
Operationally, exploitation aligns with the MITRE ATT&CK technique Masquerading (T1036); ranked at the 31th percentile by exploit likelihood (below the median); it is not currently listed in the CISA KEV catalog.
This vulnerability is AI-related — categorised as Enterprise AI Assistants.
The strongest mitigations our analysis identified map to SA-11 (Developer Testing and Evaluation) and SI-10 (Information Input Validation) — see the control section below for these in your framework.
OWASP Top 10 for Web (2025)
EU & UK References
- 🇪🇺 ENISA EUVD: EUVD-2026-51564
Vulnerability Data
The fix for CVE-2026-12045 in pgAdmin 4 9.16 required the LLM-supplied query passed to the AI Assistant's execute_sql_query tool to parse, via sqlparse, as exactly one non-transaction-control statement before running it inside a BEGIN TRANSACTION READ ONLY wrapper. sqlparse's string-literal…
more
lexing can disagree with PostgreSQL's own parser: under standard_conforming_strings = on (PostgreSQL's default since 9.1), a backslash immediately before a quote is an ordinary character to PostgreSQL, but sqlparse treats it as escaping the quote. A payload such as SELECT '\';COMMIT;CREATE TABLE pwn(x int);SELECT 1 --' therefore parses as a single SELECT to sqlparse's validator, while PostgreSQL executes it as four statements: the smuggled COMMIT ends the wrapping read-only transaction, and the trailing ROLLBACK becomes a no-op. This reintroduces the same write/RCE bypass CVE-2026-12045 was meant to close, reachable via the same indirect prompt-injection delivery (an attacker plants the payload in any object the AI Assistant may read; the LLM emits it as a tool call). An initial candidate fix ran the query with psycopg's execute(..., prepare=True), intending to force PostgreSQL's own Parse step (extended query protocol) to reject multi-statement text regardless of sqlparse's classification. This candidate fix does not work as submitted: psycopg3's PrepareManager silently ignores the prepare argument whenever the connection's prepare_threshold is None, which is pgAdmin's default for every server connection (the per-server "Prepare threshold" field is blank unless an administrator explicitly sets it) -- psycopg3 falls back to the simple query protocol, the same multi-statement-capable path the bypass exploits, so the candidate fix closes nothing on any real-world default configuration. The corrected fix sets conn.prepare_threshold = 0 directly on the dedicated, single-use read-only connection the AI Assistant tool opens, structurally forcing the extended query protocol independent of any server-level configuration. Verified against a live PostgreSQL 18 instance: the payload executes successfully under the prepare_threshold=None (default) behavior, and is rejected with "cannot insert multiple commands into a prepared statement" once prepare_threshold=0 is set on that connection. This issue affects pgAdmin 4: from 9.13 before 9.17.
- CWE(s)
AI Security AnalysisAI
- AI Category
- Enterprise AI Assistants
- Risk Domain
- N/A
- OWASP Top 10 for LLMs 2025
- None mapped
- AI-specific weaknesses CR
- CWE-1426 — Untrusted LLM output reaches SQL sink without correct validation.
Mapped by Cyber Resilience · not in NVD. Poisoning and extraction cases are routed to MITRE ATLAS instead of a synthetic CWE.- Classification Reason
- Matched keywords: llm, ai, ai, llm, ai
Related Threats
MITRE ATT&CK Enterprise Techniques
CVEs Like This One
Affected Assets
Mitigating Controls
Control response
—
—
—
V6.2.5
Mitigating Controls (NIST 800-53 r5) AI
Developer testing and evaluation can discover SQLi flaws before deployment but does not stop their introduction.
Input validation directly stops untrusted data from reaching SQL query construction without neutralization.
Secure engineering principles require parameterized queries and input sanitization that structurally eliminate SQLi.
System monitoring can identify attempted SQLi exploitation via anomalous queries after the weakness exists.
Mitigating Controls (NIST CSF 2.0) AI
Derived directly from the weakness types (CWEs) cited in the NVD entry via our AI-authored CWE→CSF cross-walk (authority under review) — links open the control.
Secure SDLC practices directly target injection flaws during coding and review so largely prevent CWE-89 introduction, yet the single broad outcome leaves residual risk from incomplete neutralization techniques or missed edge cases.
Training raises developer awareness of SQLi risks and can reduce introduction likelihood (partial) but removes none of the actual coding flaw's risk by itself since technical neutralization is still required.
Mitigating Controls (ISO/IEC 27001:2022 Annex A) AI
Derived directly from the weakness types (CWEs) cited in the NVD entry via our AI-authored CWE→ISO cross-walk (authority under review) — links open the control.
The same secure-coding and static-analysis activities surface missing neutralization of SQL metacharacters before the system is accepted.
Early warnings and shared best-practice information help organizations apply the latest remediation techniques against SQL-injection vulnerabilities.
Threat-intelligence feeds that surface new SQL-injection campaigns enable rapid updates to query-construction defenses and detection signatures before exploitation occurs.
Secure-coding rules and security testing phases mandate the use of parameterized queries or equivalent escaping, preventing the construction of dynamic SQL statements from untrusted input.
Application security requirements explicitly call for robust input handling and sanitization to avoid misinterpretation.
Secure architecture principles include defensive input processing and error handling that reduce misinterpretation risks.