Second-model AI auto-review for DeepSeek Harness approval requests: a read-only reviewer subagent returns structured allow/deny verdicts with reasons, fail-closed by default, fully auditable from the session log (approval/asked -> autoReview/verdict -> approval/decided).
Claude Code-style declarative permission rules for DeepSeek Harness: ordered allow/deny/ask rules with tool-name, argument (glob/regex), and workspace-path matching on the tools/pre-execute waterfall, session-log audit, and HMR reload.
Login gate for the DeepSeek Harness (dsh) web surface: password or shared-token authentication, session cookies, rate limiting, and a user-management CLI. | DeepSeek Harness (dsh) 网页版登录门插件:账号口令或共享令牌认证、会话 cookie、登录限速,附用户管理 CLI。
OpenAI Codex Security for DSH, zero-setup — 13 audit skills + 5 scan tools in a session preset, plus an auto-audit gate for new plugins, running on your local model with no API keys.