Beyond Attack-Success Rate: Action-Graded Severity Scale for Tool-Using AI Agents

• URL: http://arxiv.org/abs/2607.07474v1
• Date: Wed, 08 Jul 2026 14:38:01 GMT
• Title: Beyond Attack-Success Rate: Action-Graded Severity Scale for Tool-Using AI Agents
• Abstract summary: Agentic red-teaming benchmarks report whether an injected agent was compromised as a single bit: the attack succeeded, or it did not.<n>We argue that this binary attack-success rate discards the information a defender most needs, namely how harmful the resulting action was.<n>We introduce an action-graded harm rubric that scores an agent's tool-call trajectory on a seven-level ordinal scale (L0 to L6) according to whether the executed action was reversible, whether it crossed scope to reach another party, and whether it expanded privilege.
• Score: 0.0
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Agentic red-teaming benchmarks report whether an injected agent was compromised as a single bit: the attack succeeded, or it did not. We argue that this binary attack-success rate discards the information a defender most needs, namely how harmful the resulting action was. We introduce an action-graded harm rubric that scores an agent's tool-call trajectory on a seven-level ordinal scale (L0 to L6) according to whether the executed action was reversible, whether it crossed scope to reach another party, and whether it expanded privilege. We compute the scale two ways: a deterministic oracle that reads the trajectory and the attacker's stated goal, and a panel of three frontier language-model judges that read a tag-free account of the same trajectory. Across four victim models and two defenses on the AgentDojo workspace suite, severity grading exposes three cases the binary metric hides, including a defense that reports a zero attack-success rate while still permitting an externally visible cross-scope leak through an unfiltered tool. The judge panel reproduces the oracle with high ordinal agreement (Krippendorff's alpha = 0.91) but shares systematic blind spots that we characterize, most notably a failure to recognize escalation chains. Unlike prior work that provides harm taxonomies, harmful-task completion tests, execution-level safety benchmarks, or severity-aware simulation, our contribution is a reusable, trace-grounded severity instrument applied to the actual actions recorded in existing red-team logs. All code, prompts, and per-episode logs are released.

Related papers

• AgentDrift: A Step-Labeled Benchmark of Injection-Hijacked LLM Agent Trajectories [0.0]
  LLM agents complete tasks by issuing sequences of tool calls, and every observation they read is a channel through which an indirect prompt injection can enter.<n>Existing benchmarks measure whether such attacks succeed against live agents, and existing guard models judge a trace as a whole.<n>We present AgentDrift, a benchmark of 12,536 synthetic tool-call trajectories over five agent domains.
  arXiv  Detail & Related papers  (2026-09-07T03:10:23Z)
• SIR: Self-improving Red-teaming for Compute Use Agents [71.71987044117371]
  Computer use agents (CUAs) are vision-language models that perceive a screen and act on a real operating system through mouse, keyboard, and terminal.<n>Because they can be exposed to untrusted content while operating, they are vulnerable to indirect prompt injection (IPI)<n>We present SIR, a black box IPI attack that composes stealthy injections from a small library of reusable principles stated in plain language.
  arXiv  Detail & Related papers  (2026-08-31T03:39:43Z)
• ROPE: Routed Origin Policy Enforcement against Indirect Prompt Injection [61.4148196085256]
  We present ROPE (Routed Origin Policy Enforcement), which is anchored in a structural notion of trust.<n>Our approach admits two provable guarantees: 1) at every step of a trajectory, no value whose only origin is attacker-writable content reaches an origin-guarded parameter, and 2) no rewording of an injection changes an admission decision.
  arXiv  Detail & Related papers  (2026-08-27T01:22:14Z)
• Defending against Adaptive Prompt Injection Attacks via Reasoning-enabled Task Alignment [25.752599132396437]
  Indirect prompt injection attacks hijack LLM-based agents by embedding malicious instructions in third-party data that the agent retrieves during task execution.<n>Existing defenses report near-zero attack success rate on static benchmarks, yet recent adaptive evaluations show that these results collapse once the attacker is allowed to optimize against the deployed defense.<n>We propose RETA, a training-based method that grounds defense decisions on the user tasks rather than attacker-controlled data.
  arXiv  Detail & Related papers  (2026-06-13T19:15:44Z)
• AutoDojo: Adaptive Black-Box Attacks Reveal the Limits of IPI Defenses and Task-Specification Effects in LLM Agents [57.34566159148893]
  Indirect prompt injection (IPI) is a major security threat to LLM-powered agents.<n>We develop AutoDojo, an adaptive extension of AgentDojo that optimize IPI against a given defense.
  arXiv  Detail & Related papers  (2026-06-13T02:09:08Z)
• TwinGate: Stateful Defense against Decompositional Jailbreaks in Untraceable Traffic via Asymmetric Contrastive Learning [60.68349524623048]
  Decompositional jailbreaks pose a critical threat to large language models.<n>We present TwinGate, a stateful dual-encoder defense framework.<n>We construct a comprehensive dataset of over 3.62 million instructions spanning 8,600 distinct malicious intents.
  arXiv  Detail & Related papers  (2026-04-30T13:44:01Z)
• PlanGuard: Defending Agents against Indirect Prompt Injection via Planning-based Consistency Verification [1.7904458681854372]
  PlanGuard is a training-free defense framework based on the principle of Context Isolation.<n>PlanGuard effectively neutralizes attacks, reducing the Attack Success Rate (ASR) from 72.8% to 0%.
  arXiv  Detail & Related papers  (2026-04-11T09:59:46Z)
• Visual Confused Deputy: Exploiting and Defending Perception Failures in Computer-Using Agents [9.457255218406333]
  Computer-using agents (CUAs) act directly on graphical user interfaces, yet their perception of the screen is often unreliable.<n>We formalize the visual confused deputy: a failure mode in which an agent authorizes an action based on a misperceived screen state.<n>To mitigate this threat, we propose the first guardrail that operates outside the agent's perceptual loop.
  arXiv  Detail & Related papers  (2026-03-16T01:31:32Z)
• SkillJect: Automating Stealthy Skill-Based Prompt Injection for Coding Agents with Trace-Driven Closed-Loop Refinement [120.52289344734415]
  We propose an automated framework for stealthy prompt injection tailored to agent skills.<n>The framework forms a closed loop with three agents: an Attack Agent that synthesizes injection skills under explicit stealth constraints, a Code Agent that executes tasks using the injected skills and an Evaluate Agent that logs action traces.<n>Our method consistently achieves high attack success rates under realistic settings.
  arXiv  Detail & Related papers  (2026-02-15T16:09:48Z)
• CausalArmor: Efficient Indirect Prompt Injection Guardrails via Causal Attribution [49.689452243966315]
  AI agents equipped with tool-calling capabilities are susceptible to Indirect Prompt Injection (IPI) attacks.<n>We propose CausalArmor, a selective defense framework that computes lightweight, leave-one-out attributions at privileged decision points.<n> Experiments on AgentDojo and DoomArena demonstrate that CausalArmor matches the security of aggressive defenses.
  arXiv  Detail & Related papers  (2026-02-08T11:34:08Z)
• A Coin Flip for Safety: LLM Judges Fail to Reliably Measure Adversarial Robustness [57.510025257780306]
  We show that existing validation protocols fail to account for substantial distribution shifts inherent to red-teaming.<n>We propose ReliableBench, a benchmark of behaviors that remain more consistently judgeable, and JudgeStressTest, a dataset designed to expose judge failures.
  arXiv  Detail & Related papers  (2026-02-04T15:13:35Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.