A pre-execution guard for AI coding agents. It blocks destructive Git and file system commands, plus common attempts to access sensitive files, before a tool call runs. Supports Amp Code, Antigravity CLI, Claude Code, Codex, Cursor, DeepSeek Harness, Gemini CLI, GitHub Copilot CLI, Grok Build, Hermes Agent, Kimi Code, OpenClaw, OpenCode, and Pi.
Render Mermaid code blocks as SVG diagrams in DeepSeek Harness Web, with diagram/code switching, streaming support, dark mode, strict security, and npm/GitHub installation.
Trust pipeline for deepseek-harness plugins: deterministic static scan (11 rules) + LLM-driven audit protocol + two-part scorecard, with optional runtime guard (T1 sentinel / T2 fs & child_process hooks) and honeypot decoys. Monitor-and-alert only — vet never blocks or kills on its own.
基于 better-sidebar 的统一引用插件：划选文件/消息即可批注引用，引用块带文件路径与行区间、双端折叠、点击跳回原文 ｜ Unified-quoting plugin built on DSH-better-sidebar: select files or messages to annotate; quote blocks carry path & line range, collapse in chat and history, and jump back to the source on click.
Enforced pain-point-check guard plugin for DeepSeek Harness: after two non-converged experiments it injects the three questions, denies non-investigative tool calls until answered, and blocks same-direction retries. dsh-plugin
Enforce strict checks as dsh harness policy: repeated-failure intervention, a glob-protected critical-path gate that blocks with diagnostics, and an automatic post-write check.
DSH (DeepSeek Harness) plugin that hides or blocks sensitive filesystem paths from AI agents — privacy, sandboxing, guardrails & human-in-the-loop approval for Node.js
