# The Guard That Cried Wolf: How Scary Words Make Agent Guardrails Refuse Legitimate Actions

arXiv: https://arxiv.org/abs/2608.27009 · HTML: https://mlsanity.com/paper/2608.27009/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Placed in Agent threats, guardrails and governance, one of 16 directions in LLM agents and tool use.
- Builds on 16 of its 26 references that are in the map.
- First posted on arXiv 27 Aug 2026.

## Placed in

- LLM agents and tool use / [Agent threats, guardrails and governance](https://mlsanity.com/agents/agent-threats-guardrails-and-governance/)
- LLM safety and security / [Prompt injection and agent security](https://mlsanity.com/safety/prompt-injection-and-agent-security/)

## Builds on (most cited first, 16 of 16)

- [Llama Guard: LLM-based Input-Output Safeguard for Human-AI Conversations](https://mlsanity.com/paper/2312.06674/) (2023)
- [Identifying the Risks of LM Agents with an LM-Emulated Sandbox](https://mlsanity.com/paper/2309.15817/) (2023)
- [WildGuard: Open One-Stop Moderation Tools for Safety Risks, Jailbreaks, and Refusals of LLMs](https://mlsanity.com/paper/2406.18495/) (2024)
- [The Instruction Hierarchy: Training LLMs to Prioritize Privileged Instructions](https://mlsanity.com/paper/2404.13208/) (2024)
- [Measuring what Matters: Construct Validity in Large Language Model Benchmarks](https://mlsanity.com/paper/2511.04703/) (2025)
- [OS-Harm: A Benchmark for Measuring Safety of Computer Use Agents](https://mlsanity.com/paper/2506.14866/) (2025)
- [System-Level Defense against Indirect Prompt Injection Attacks: An Information Flow Control Perspective](https://mlsanity.com/paper/2409.19091/) (2024)
- [SafetyPrompts: a Systematic Review of Open Datasets for Evaluating and Improving Large Language Model Safety](https://mlsanity.com/paper/2404.05399/) (2024)
- [ToolSafe: Enhancing Tool Invocation Safety of LLM-based agents via Proactive Step-level Guardrail and Feedback](https://mlsanity.com/paper/2601.10156/) (2026)
- [Safetywashing: Do AI Safety Benchmarks Actually Measure Safety Progress?](https://mlsanity.com/paper/2407.21792/) (2024)
- [Poison Once, Exploit Forever: Environment-Injected Memory Poisoning Attacks on Web Agents](https://mlsanity.com/paper/2604.02623/) (2026)
- [Temporary Authority, Permanent Effects: Commit-Time Authorization for LLM Agents](https://mlsanity.com/paper/2607.10487/) (2026)
- [A Framework for Formalizing LLM Agent Security](https://mlsanity.com/paper/2603.19469/) (2026)
- [When Benign Inputs Lead to Severe Harms: Eliciting Unsafe Unintended Behaviors of Computer-Use Agents](https://mlsanity.com/paper/2602.08235/) (2026)
- [Establishing Construct Validity in LLM Capability Benchmarks Requires Nomological Networks](https://mlsanity.com/paper/2603.15121/) (2026)
- [SentinelAgent: Intent-Verified Delegation Chains for Securing Federal Multi-Agent AI Systems](https://mlsanity.com/paper/2604.02767/) (2026)

## Abstract (arXiv)

Agent guardrails are checks that approve or refuse each action before an LLM executes it. Sometimes they refuse requests that are genuinely safe. This over-safety blocks deployment when a guardrail refuses an authorized task. Evaluating over-safety is hard: at the boundary an authorized action resembles an unauthorized one, and the safe-versus-unsafe label is a choice of authorization policy, not fixed by the action alone. We argue it therefore requires a benchmark that does not yet exist, one that maps the decision boundary of an ideal guardrail. Harvesting such a benchmark from real data is impractical: boundary cases are hard to collect, their labels hard to verify. The gap is real, so we construct Cautious Bench, the first benchmark to make over-safety the construct for agent guardrails; it codesigns each sample and its label with a stated authorization policy. A build-time gate re-derives every example to certify it, so each label is a mechanical consequence of the policy rather than an annotator's per-sample verdict, a reference against which researchers can measure real guardrails. The benchmark renders 756 Decidable benign/twin pairs, each under three object-name types (2,268 measured pairs), and 40 Undecidable pairs reported separately. Measuring six guardrails from five designs, we find a name-superstition effect: each over-refuses an authorized action more often under a scary-looking object name than a benign one. Since only the object name varies in the aforementioned contrast experiments, the deviation is the name's doing: the guardrails read the surface label, not the authorization context.