dsh-taintguard
当工具结果包含不可信内容时将代理标记为已污染，对随后的高权限调用进行拦截，并在所有模式下拒绝将凭据传递给可联网的工具。
Taints the agent when tool results carry untrusted content, gates the privileged calls that follow, and refuses credentials passed to network-capable tools in every mode.
怎么安装
dsh plugin add github:sashankh/dsh-taintguard 简介
dsh-taintguard Indirect prompt-injection guard for DeepSeek Harness. An agent that browses the web or reads a repository puts third-party text into the same channel that carries its operator's instructions. A model has no reliable way to tell them apart, so a page, a README, or an MCP tool result can ask the agent to run a command or post a file somewhere. This plugin does not try to make the model immune to that. It narrows what the retrieved text can reach: it marks the agent once a tool result arrives from an untrusted source, gates the privileged calls that follow, and refuses outright whe…
推荐参考
信息
- 协议
- MIT
- 语言
- TypeScript
- GitHub 星标
- 0
- 月下载
- –
- 最近更新
- 2026-08-16
- 创建于
- 2026-08-16
基础安全检查
- 检查结果
- curated 收录但无 npm 包/安装命令
- 收录来源
- curated:awesome-dsh-plugin.com, curated:awesome-dsh-plugin/awesome-dsh-plugin
- 主题标签
- agent-security, ai-security, deepseek-harness, dsh, dsh-plugin, guardrails, prompt-injection
同类推荐
dsh-auto-review
PerryLink/dsh-auto-review
审批链上的第二模型自动审查：只读审查子代理返回带理由的 allow/deny 结构化裁决，默认 fail-closed。
dsh-passwords
slywalker2006/dsh-passwords
让 DeepSeek Harness 变成服务器级多租户平台：远程访问 + 自动 HTTPS、子用户权限与配额、沙盒强制、加密认证与审计日志。
dsh-auth-gate
TecFancy/dsh-auth-gate
DSH 网页端登录门插件：账号口令或共享令牌认证、会话 cookie、登录限速，附用户管理 CLI（0.4.1 起声明 dsh.bundle manifest，`dsh plugin add` 一键挂载）。
dsh-approval-llm
Letter2025/dsh-approval-llm
基于模型的权限审批：由独立审查模型自动应答 approval 权限请求。