Ten major security mechanisms of 40 Coding Agents
Sun Yat-sen University, Peng Cheng Laboratory and other institutions released the HarnessSecurity-Bench paper in October 2026, investigating 40 Coding Agents and summarizing ten major security mechanisms, then selecting 6 products to conduct 2,500 experiments across 23 programming tasks with GLM-5.2 as the unified base model. Results show that only 4 — Claude Code, Gemini CLI, gptme and Goose — could be confirmed to implement prompt injection filtering; after enabling automatic authorization, task utility rose from 77.1% to 95.4%, while attack success rate rose from 29.2% to 95.6%; network isolation reduced attack success rate from 57.1% to 0.8%, but task utility dropped by 24.5 percentage points. The paper also uses the Qwen Code case to show that a command whitelist can be bypassed by an allowed Python interpreter, and the article accordingly recommends that enterprises check the default enabled status of security mechanisms and evaluate security, task utility and execution cost together.
The paper incorporates the security mechanisms of 40 Coding Agents into a unified taxonomy and empirically measures the utility and attack changes before and after toggling them, so that enterprises evaluating Agents can use it to check default configurations against actual protections.