OWASP Agentic
ASI01 · Agent Goal Hijack
Attackers redirect an agent's objectives or plan, typically through injected content in documents, mail or web pages.
OWASP Top 10 for Agentic Applications (2026)
Mitigations
- Isolate untrusted content from planning
- Re-validate goals before sensitive steps
- Monitor for plan deviations
Related techniques
- LLM01 Prompt Injection OWASP LLM
- AML.T0051 LLM Prompt Injection MITRE ATLAS
In the news
-
Kaer: Security-Aware Replanning for Multi-Agent Systems
Kaer is a multi-agent framework that shields the planner from untrusted agent feedback to prevent instructions embedded in feedback from hijacking replanning.
ReviewReview multi-agent orchestrator designs for unsanitized agent feedback reaching the planner
-
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
Tools
Test or exercise it
- promptfoo LLM red-teaming · Adopt · Dual-use
Detect or mitigate it
- AgentDojo Benchmark · Trial
- beelzebub Benchmark · Assess
- promptfoo LLM red-teaming · Adopt
- PurpleLlama (Llama Guard, Prompt Guard, LlamaFirewall) Guardrail · Trial
Mappings are AI-assisted and checked against the framework definitions; corrections welcome.