postgres-ai-tenant-boundary-v2
7 rules · Keryn Software Licence · rules 0.8.0 · all packs
How to read a rule
- Severity
- How bad the finding is if it is real:
errorfails a default gate,warningandinfodo not. - Confidence
- P0 exploitable as written; P1 dangerous in context; P2 a heuristic or a posture check: confirm by reading the code.
- False-positive risk
- low: an exact API or flag, so hits are almost always real; medium: the pattern is legitimate in many programs, so check each hit; high: it finds a lead to confirm, so expect benign hits.
- Edge
- The relationship the finding sits on. reaches: content or traffic gets to a thing; calls: one thing invokes another; reads: one thing reads from another; writes: one thing writes to another; sends: data leaves for another party; grants: authority is given or left unchecked; depends_on: something relies on a component it does not verify; trusts: one party believes another without evidence. The boundary it crosses, when there is one: user->prompt (untrusted input reaching a prompt); model->code (model output reaching code, a shell, a query or a page); secret->outside (a secret leaving the trusted side); supply-chain (code, models or prompts pulled in from outside).
- Taxonomy chips
- References for readers who work in OWASP LLM 2025, OWASP Agentic, CWE or NIST AI RMF. They say where a rule belongs, not that a framework is satisfied.
seceng.postgres.pgvector-query-without-tenant-filter
Section titled “seceng.postgres.pgvector-query-without-tenant-filter”warning python, typescript, javascript, go, rust postgres
pgvector similarity query over AI/RAG data lacks a visible tenant, user, workspace, or organization predicate.
Confidence P2: a heuristic or a posture check: confirm by reading the code. False-positive risk medium: the pattern is legitimate in many programs, so check each hit.
Edge inspected: grants authority is given or left unchecked.
Also reported under the older id seceng.neon-vector-query-without-tenant, seceng.supabase.vector-similarity-query-without-tenant-predicate.
seceng.postgres-ai-table-without-rls
Section titled “seceng.postgres-ai-table-without-rls”warning generic postgres
Postgres table containing AI chat, memory, prompt, document, embedding, or agent-run data is created without a nearby RLS enablement statement.
Confidence P1: dangerous in context. False-positive risk high: it finds a lead to confirm, so expect benign hits.
Edge inspected: grants authority is given or left unchecked.
CWE-862 LLM02 LLM08 ASI03 ASI06 ASI02
seceng.postgres-security-definer-rag-function
Section titled “seceng.postgres-security-definer-rag-function”warning generic postgres
SECURITY DEFINER function touches AI/RAG tables. Verify tenant scoping, search_path, and caller authorization because this can bypass RLS.
Confidence P1: dangerous in context. False-positive risk high: it finds a lead to confirm, so expect benign hits.
Edge inspected: reads one thing reads from another.
seceng.postgres-vector-rpc-missing-auth-uid
Section titled “seceng.postgres-vector-rpc-missing-auth-uid”warning generic postgres
Vector search SQL function appears to lack auth.uid(), tenant_id, user_id, workspace_id, or organization_id checks.
Confidence P2: a heuristic or a posture check: confirm by reading the code. False-positive risk medium: the pattern is legitimate in many programs, so check each hit.
Edge inspected: grants authority is given or left unchecked.
seceng.postgres-ai-function-dynamic-sql
Section titled “seceng.postgres-ai-function-dynamic-sql”warning generic postgres
AI/RAG SQL function uses dynamic SQL. Dynamic SQL around vector or prompt data can create injection or authorization-bypass risk.
Confidence P1: dangerous in context. False-positive risk high: it finds a lead to confirm, so expect benign hits.
Edge inspected: calls one thing invokes another; it crosses the model->code boundary (model output reaching code, a shell, a query or a page).
CWE-89 LLM08 LLM05 ASI01 ASI03 ASI06
seceng.postgres-prompt-or-secret-column-in-public-view
Section titled “seceng.postgres-prompt-or-secret-column-in-public-view”warning generic postgres
Database view appears to expose prompt, system instruction, tool, secret, or transcript columns.
Confidence P1: dangerous in context. False-positive risk high: it finds a lead to confirm, so expect benign hits.
Edge inspected: writes one thing writes to another.
CWE-200 LLM02 LLM07 ASI03 ASI02 ASI10
seceng.postgres-ai-audit-log-raw-prompt
Section titled “seceng.postgres-ai-audit-log-raw-prompt”info python, typescript, javascript postgres
SQL insert or table schema appears to store raw prompt/messages/completions in audit logs without a redaction hint.
Confidence P1: dangerous in context. False-positive risk high: it finds a lead to confirm, so expect benign hits.
Edge inspected: writes one thing writes to another.