Permission Denied: Policy-Graded Evaluation of Coding Agents in Hardened Environments

arXiv:2608.02670 · cs.CR, cs.AI · Submitted 2026-08-02 · Read on arXiv

Dotan Davidovich, Yair Amar, Hai Rozencwajg, Or Hiltch

cs.CR, cs.AI

Submitted: 2026-08-02

Code: https://github.com/boundary-bench/boundary-bench

Project page: https://boundarybench.com

License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/

Terminology

• Verifier — 62× in this paper · explained in 2 episodes
• Harness — 49× in this paper · explained in 2 episodes
• Benchmark — 39× in this paper · explained in 2 episodes
• Inflation — 11× in this paper · explained in 2 episodes
• Pareto Frontier — 11× in this paper · explained in 2 episodes
• Trace — 9× in this paper · explained in 2 episodes
• Robustness — 8× in this paper · explained in 6 episodes
• Ablation — 4× in this paper · explained in 1 episode
• Absorb — 4× in this paper · explained in 1 episode
• AI Agents — 3× in this paper · explained in 2 episodes
• Benchmarking — 3× in this paper · explained in 4 episodes
• Provenance — 3× in this paper · explained in 5 episodes
• Verification — 3× in this paper · explained in 2 episodes
• Agent Harness — 2× in this paper · explained in 2 episodes
• Core — 2× in this paper · explained in 2 episodes
• Ablation Study — 1× in this paper · explained in 3 episodes
• Agentic System — 1× in this paper · explained in 2 episodes
• Decoupling — 1× in this paper · explained in 2 episodes
• Distribution Shift — 1× in this paper · explained in 12 episodes
• Distribution Shifts — 1× in this paper · explained in 2 episodes
• Failure Modes — 1× in this paper · explained in 3 episodes
• Memorization — 1× in this paper · explained in 4 episodes
• Tokenization — 1× in this paper · explained in 2 episodes

Sources

• ST-WebAgentBench: A Benchmark for Evaluating Safety and Trustworthiness in Web Agents
• Terminal-Bench: Benchmarking Agents on Hard, Realistic Tasks in Command Line Interfaces
• Where Do LLMs Still Struggle? An In-Depth Analysis of Code Generation Benchmarks
• More Vulnerable than You Think: On the Stability of Tool-Integrated LLM Agents
• $\tau$-bench: A Benchmark for Tool-Agent-User Interaction in Real-World Domains

Related papers

• SoK: AI-Augmented Binary Reversing
• Relaxed Sender Anonymity for CBDC Interbank Settlement: A Zero-Knowledge Approach on Permissioned EVM
• Calibration-Family Overfit: Why Trusted Sabotage Monitors Don't Transfer Across Lineages
• Estimating the Social Cost of Corporate Data Breaches
• Efficient Fuzzy PSI under One-Sided Assumptions
• Sealing the Audit-Runtime Gap for LLM Skills