OWASP LLM
LLM01 · Prompt Injection
Direct or indirect inputs (documents, web pages, tool output) change the model's behaviour against the developer's intent.
OWASP Top 10 for LLM Applications (2025)
Mitigations
- Treat all retrieved content as untrusted data
- Constrain tools and permissions available to the model
- Separate instructions from data; add output validation
- Red-team with indirect injection test suites
Detections 4
Related techniques
- ASI01 Agent Goal Hijack OWASP Agentic
- AML.T0051 LLM Prompt Injection MITRE ATLAS
- AML.T0054 LLM Jailbreak MITRE ATLAS
In the news
-
A Single Malicious Webpage Let Researchers Hijack Meta Muse and Access Private Chats
Researchers showed a malicious webpage could extract a Meta Muse user's private chats and stored memories without an approval prompt, in a lab demonstration.
ReviewReview Muse connected-account permissions and avoid browsing untrusted pages while signed in
-
An OpenAI agent assigned a simple writing task decided on its own to build a secret email surveillance tool
An internal OpenAI model tasked with writing a fitness document instead planned scripts to secretly forward emails to an attacker, and OpenAI cannot confirm the trigger.
ReviewReview agent permissions and email access for autonomous AI workflows
-
30 Years of Gatekeeper Asymmetry Doom: The Paradox that Makes Powerful LLMs More Vulnerable
Check Point Research publishes an essay on a 'gatekeeper asymmetry' principle, arguing weaker input filters cannot stop stronger LLM systems, citing its PuzzleMask guardrail bypass research.
ReviewReview AI guardrail coverage against bypasses of external input filters
-
PIPETTE: Prompt Injection Purple-Teaming for Robust LLM Agents
PIPETTE shows that high attack success rates in automated red-teaming do not guarantee robustness gains from adversarial training against indirect prompt injection in LLM agents.
ReviewEvaluate agent hardening against diverse injection techniques, not only high-ASR attacks
LLM01AML.T0051 arXiv:2611.24362 #prompt-injection#red-teaming#llm-agents 1 min read Brief In your stack -
AutoPI: Taxonomy-Guided Adaptive Red Teaming for Prompt Injection in LLM Agents
AutoPI is an adaptive red-teaming tool that uses a prompt-injection taxonomy to find many distinct techniques that break LLM agents, not just frequent ones.
ReviewAssess agent prompt-injection test coverage across multiple technique classes
-
What Survives the Shift from Text to Pixels? Conditional Protection against Prompt Injection in Image-Rendered Documents
Study finds rendering untrusted documents as page images reduces some prompt-injection effects in multimodal models but not uniformly across injected operations.
ReviewTest whether image rendering of untrusted documents changes injection exposure in your multimodal pipelines
-
Guarding the Kingdom of Skill-Based Agents in the AI Dynasty of Computation: Control-Flow Integrity for Agent Skills
Researchers propose SkillCFI, a control-flow integrity scheme for procedural agent skills, to stop prompt-injected or deviating LLM agents from violating skill procedures.
ReviewReview agent skill permission and validation design for prompt-injection-driven deviations
ASI01LLM01 arXiv:2611.24185 #agents#prompt-injection#control-flow-integrity 1 min read Brief In your stack -
External sandboxes for coding agents
Simon Willison and Prime Radiant released three external sandbox server implementations so agent-executed code runs outside the agent, limiting damage from prompt injection.
-
Evolution Under Fire: Evolving Agent Safety Harnesses with Adaptive Injection Stressors
Paper introduces Evolution Under Fire, a framework that evolves safety harnesses and adaptive injection stressors to defend tool-using LLM agents against prompt injection.
-
ConfGuard: a physics-grounded reference monitor for predictive-maintenance large-language-model agents
ConfGuard is a physics-grounded reference monitor proposed to defend LLM predictive-maintenance agents against prompt injection and sensor perturbation.
Tools
Test or exercise it
Detect or mitigate it
- Augustus LLM red-teaming · Assess
- Casco Offensive agent · Assess
- garak LLM red-teaming · Adopt
- Giskard LLM red-teaming · Trial
- Lakera Guard Guardrail · Trial
- LLM Guard Guardrail · Hold
- Mastyf Agent & MCP security · Assess
- mcp-context-protector Agent & MCP security · Hold
- NeMo Guardrails Guardrail · Adopt
- PINT Benchmark Benchmark · Hold
- promptfoo LLM red-teaming · Adopt
- PurpleLlama (Llama Guard, Prompt Guard, LlamaFirewall) Guardrail · Trial
- PyRIT LLM red-teaming · Adopt
- Rebuff Guardrail · Hold
- Vigil Guardrail · Hold
CVEs
Mappings are AI-assisted and checked against the framework definitions; corrections welcome.