method · first seen 6 Oct 2026
agent guardrails
Also: agent guards, PreToolUse hooks, agent safety rules, AI agent guardrails, LLM agent guardrails, prompt injection guardrails, guardrails for language models
Components that screen an agent's proposed actions or incoming messages to allow or block them.
Mentions per week
PapersCodeDiscussion
Earliest sources
- POLAR: Ontology-Guided Risk Prevention for Tool-Calling LLM AgentsPapers · 6 Oct 2026 · Yunju Kang, Seonghyeon Cho, Irene Li, Yeo-Chan Yoon, Chanjun Park
- Formal Runtime Verification for Tool-Using LLM Agents: An Offline Same-Benchmark Study on AgentDojo and STACPapers · 7 Oct 2026 · Nikolaos Kekatos, Stylianos Basagiannis, Marinelio Chintri, Alexios Lekidis, Tom…
- One Word Opens the Gate: The Option-Channel Attack on Typed Decision Models as Agent GuardrailsPapers · 8 Oct 2026 · Seyedarmin Azizi, Erfan Baghaei Potraghloo, Massoud Pedram
- VEPA: Verified Evolving Formal Policy for Lifelong LLM Agent GuardrailsPapers · 8 Oct 2026 · Duy Nguyen, Lesly Miculicich, Bhavana Dalvi Mishra, Shuo Li, Krishnamurthy Dj Dv…
- Same Operation, Different Permission: Evaluating Whether Guardrails Honor AuthorizationPapers · 12 Oct 2026 · Huihao Jing, Shaojin Chen, Wenbin Hu, Yuxuan Liu, Haochen Shi, Zhongwei Xie, Hao…
- dooly6276/claude-bulkheadCode · 13 Oct 2026 · dooly6276 · 35 stars
Latest sources
- anyu0804/antigravity-agent-rulesCode · 9 Nov 2026 · anyu0804 · 30 stars
- GuardrailAgent: An Agentic Framework for Jailbreak DefensePapers · 6 Nov 2026 · Tianze Chang, Junjie Chu, Yixin Wu, Michael Backes, Yang Zhang
- onlyreadai/onlyreadCode · 5 Nov 2026 · onlyreadai · 93 stars
- From Clinical Judgment to Checkable Rules: Clinician-Led Guardrails for Mental-Health ChatbotsPapers · 22 Oct 2026 · Chenhan Lyu, Kit Wislocki, Sarah Elizabeth Stoeckl, Maitreyi Sinha, Amir Rahmani
- MAS-Guard: Unified Guardrails from Individual Agents to Multi-Agent LLM SystemsPapers · 20 Oct 2026 · Mohammad Kachuee, Phu Mon Htut, Sheng Li, Nikolaos Pappas, Miguel Ballesteros, Y…
- SteelUnion/dsh-confirm-gateCode · 17 Oct 2026 · SteelUnion · 101 stars
- Language Is a Security Surface for Agent GuardsPapers · 15 Oct 2026 · Hasan Demirkiran, Jens Ernstberger
- Shortcut Sensitivity in Agent-Safety EvaluationsPapers · 15 Oct 2026 · Hasan Demirkiran, Jens Ernstberger
- dooly6276/claude-bulkheadCode · 13 Oct 2026 · dooly6276 · 35 stars
- Same Operation, Different Permission: Evaluating Whether Guardrails Honor AuthorizationPapers · 12 Oct 2026 · Huihao Jing, Shaojin Chen, Wenbin Hu, Yuxuan Liu, Haochen Shi, Zhongwei Xie, Hao…
- VEPA: Verified Evolving Formal Policy for Lifelong LLM Agent GuardrailsPapers · 8 Oct 2026 · Duy Nguyen, Lesly Miculicich, Bhavana Dalvi Mishra, Shuo Li, Krishnamurthy Dj Dv…
- One Word Opens the Gate: The Option-Channel Attack on Typed Decision Models as Agent GuardrailsPapers · 8 Oct 2026 · Seyedarmin Azizi, Erfan Baghaei Potraghloo, Massoud Pedram