FengLingYaaa/dsh-gpt-perm-strip

Name: FengLingYaaa/dsh-gpt-perm-strip
Author: FengLingYaaa

Plugin ⭐ 0 MIT security tools

A DeepSeek Harness plugin that runs only for GPT-family models. Before a tool body hits DSH's sandbox escalation check, it removes sandboxpermissions / justification that are not strictly wider than the current session.

catalog description：DSH plugin: strip leftover GPT sandbox_permissions that are not strictly wider than the current session.

↗ View on GitHub 📄 Read full README

Install

sandbox escalation to "danger-full-access" is not strictly wider than this call's current "danger-full-access" mode

Add FengLingYaaa/dsh-gpt-perm-strip to your DSH config (web profile) to enable.

README

2.5 KB · Cached 2026-08-18 17:12:06 · 2.5 KB

Expand full README Collapse

README is empty.

Rendered from README.md of FengLingYaaa/dsh-gpt-perm-strip. View original on GitHub →

Install

README

Related Plugins