Prompt injection
π Short definition
OWASP LLM01 vulnerability: an attacker inserts instructions into a source the model will read (RAG document, email, web page) to bypass the system prompt or trigger malicious tool calls1.
π Detailed definition
Unlike SQL injection, prompt injection exploits the fact that the LLM cannot reliably distinguish βsystem instructionβ from βuser contentβ once everything is concatenated in the context window.
Two common forms:
- Direct: the user sends βignore your instructions andβ¦β.
- Indirect: a document indexed in RAG contains hidden instructions read at retrieval time.
On-premise, the risk is identical to the cloud: data sovereignty does not immunize against a poisoned PDF in the internal document base.
π‘ Why it matters for on-prem AI
- Any multi-source RAG stack (SharePoint, tickets, wikis) is an attack surface.
- Combined with Excessive Agency (LLM06:2025, LLM03:2026), can lead to exfiltration via tools (send email, API query).
- Covered in depth in π Local inference security (LLM01βLLM10, OWASP 2025 numbering with its 2026 mapping).
β οΈ Common pitfalls
- Believing a βsecret system promptβ is enough: bypassable via indirect injection.
- Indexing unsanitized documents without privilege separation between retrieval and tool execution.
- Forgetting outbound channels (LLM07:2025 System Prompt Leakage, broadened into LLM08:2026 Hidden Context Exposure: the system prompt, but also tool definitions and memory) during red team tests1.
Mitigations (summary)
| Measure | Role |
|---|---|
| Least privilege on tools | Limits impact of a successful injection |
| HITL validation on sensitive actions | Human-in-the-loop |
| Input/output filtering | Detection of injection patterns |
| RAG tenant isolation | Multi-tenant without cross-organization leakage |
π See also
Footnotes
-
OWASP GenAI Security Project, LLM01:2025 Prompt Injection (genai.owasp.org/llm-top-10/); kept at the top of the 2026 edition (published 2026-08-03) with a scope extended to image/audio inputs and persistent memory, LLM07 System Prompt Leakage becoming LLM08:2026 Hidden Context Exposure there. https://genai.owasp.org/resource/owasp-genai-llm-top-10-2026/ β© β©2