Security DeepSeek Harness Plugins

Open-source CMA-compatible agent runtime for any model, with MCP tools, sandboxed sessions, audit, replay, and a local console. Includes a native DeepSeek Harness bundle over stdio MCP.

Open-source sandboxes for AI agents, untrusted code execution, and durable services.

a coding Agent from pi. ∞ providers, sub-agents, hashline edits, and a permission gate

deepseek harness对话和代码状态回退插件 | DSH — rewind conversation and workspace state, powered by a persistent Change Ledger

The vendor-neutral protocol for AI agent safety & security — and the neutral benchmark that ranks the vendors.

Create and manage sandboxed JavaScript tools for DeepSeek Harness with a Monaco editor and model-driven tool lifecycle.

为 DeepSeek Harness 提供电脑控制插件：新鲜 Accessibility 观测、过期状态拒绝、作用域权限与安全输入（目前支持macos）｜Accessibility-first macOS Computer Use bundle for DSH with fresh observations, stale-state rejection, scoped permissions, and safe input.

DSH 本机安全审计插件：配置/插件来源/会话/网络暴露面，只读脱敏风险报告

An agent-native MCP for composing, tuning, rendering, mixing, and auditing native VOCALOID3/4 projects — built just for fun.

Role-based Codex / Claude Code / ACP subagent providers for the DeepSeek Harness — continuable children, durable session recovery, per-role product permissions, and delegation with a permission ceiling.

DSH 上下文注入审计插件：统计 AGENTS.md 指令链/技能目录/工具 schema 的 token 成本，检测重复与冲突；Web UI 圆环面板 + context_audit 工具。Context Doctor for DeepSeek Harness: audit instruction-chain / skill catalog / tool schemas token cost.

CloserAI - a local-first, model-agnostic, permission-transparent desktop AI workbench built on DeepSeek Harness.

Safe OpenAPI 3.x discovery and API calling tools for DeepSeek Harness

A deepseek harness plugin for tensorlake sandbox

DSH Session Manager: manage conversations, archive/restore, delete safely, open record folders.

Built-in plugin catalog and safe GUI toggles for DeepSeek Harness Web.

DSH Web 广告拦截器，四层独立防御拦截 dsh-ads 插件的所有广告位 | DSH Web ad blocker with four independent defense layers targeting the dsh-ads plugin

Safe delete plugin for DeepSeek Harness (DSH): move files to trash / staging area instead of permanent removal, with restore and purge support.

模式匹配自动注入 steering rules，不占系统上下文 - Inject rules when needed, without wasting context. Similar to oh-my-pi's "Time-traveling stream rules", but with a very simple and compact code implementation.

A curated, evidence-led directory of DeepSeek Harness (DSH) plugins: verified loadable extensions, skills, and permission-aware installation guidance.

DeepSeek Harness plugin for safely discovering, auditing, and adopting external Agent Skills — prompt-injection and AgentBaiting defense.

Content-addressed audit indexes across independent DeepSeek Harness evidence producers

Conservative auto-approval preset for DeepSeek Harness sandbox escalations

DSH knowledge-base plugin: build audit-able KB packs (references + SQLite FTS5) from md/txt/docx/pdf, deterministic retrieval (kb_query) and original-text reading (kb_read), zero-script generated skills. Apache-2.0.

Cited memory over DSH's lossless session log — distilled, human-auditable facts with citations back to the exact source events; memory_read/memory_expand tools, recall index, and a dsh-memory CLI.

Security audit for DeepSeek Harness plugins: static permission profile with file/line evidence + a runtime sentinel gating credential access and unknown-host egress · DSH 插件安全审计：静态权限画像（附文件/行号证据）+ 运行时哨兵，触及凭证或向未知主机外发数据时先请你批准

Content-addressed, value-redacted policy drift evidence for DeepSeek Harness

Session execution analytics and audit reports for DeepSeek Harness — see how your agent actually worked

Customize the visible DSH Web thinking status with lifecycle-safe CSS.

Tiered auto-review for DeepSeek Harness: static-rule safety net + LLM reviewer + human fallback — auto-allow safe actions, deny irreversible ones, ask a human for the rest.

No repository or directory purpose description was collected.

Steam Workshop style plugin store for DeepSeek Harness Web UI: browse 850+ community plugins, one-click install with GitHub mirror acceleration, progress UI, security checks, Chinese descriptions

Observe-only hash-chained evidence receipts for DeepSeek Harness

microsandbox支持

nono沙盒支持

SecurStack adapter for DeepSeek Harness: run repository security scans, policy gates, doctor diagnostics, and JSON CLI results from safe AI-agent tools.

dsh-blocker-notify — Real-time attention alerts for DeepSeek Harness: a global banner + flashing workspace entries when the agent is blocked (approval request / sandbox denial).

Code-audited, progressive guide to production-grade DeepSeek Harness plugins

Agentic Control Plane for DeepSeek Harness — policy-check every tool call before it runs

AI approval answerer for DeepSeek Harness (DSH) using the unified LLM route with fail-closed policy checks.

Model-based permission approval (approve-for-me) for DeepSeek Harness: an approval/request answerer backed by a separate reviewer model

Trusted Auto: a middle permission tier for DeepSeek Harness between workspace-write and danger-full-access, auto-approving harmless commands and trusted-area targets

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 /rewind for DeepSeek Harness — git-first workspace snapshots before every mutation, turn-boundary session forks, one-shot /rewind restore. A dsh-plugin capability seam.

Planned OS-backed credential provider for DeepSeek Harness

Authorized security-assessment profile for DeepSeek Harness: scoped network tools, container-backed shell, authorization guard, durable evidence, 21 security skills, 7 specialist subagents

dsh-delegate: model-aware subagent delegation for DeepSeek Harness — per-call models, depends_on dependency gating, per-child personas, a durable run roster, audit events, and conversation-flow tool cards. | 给 DeepSeek Harness 的子代理委派加上：按次选模型、依赖门控、角色人设、任务花名册。

DeepSeek Harness 事件流审计面板插件：观察事件类型/分发模式/计数/最近事件，帮助插件作者理解 harness 内部

A safe, live plugin marketplace for DeepSeek Harness

Bounded, layered, approval-gated, auditable cross-session memory for DeepSeek Harness (capability seam: ctx.memory + SQLite provider + memory tool + frozen snapshot injection)

Multi-tenant SaaS extension for DeepSeek Harness (DSH): tenant identity, session isolation, authorization, tenant-aware MCP, and audit.

dsh-passwords: DeepSeek Harness login gateway - first-run setup, at-rest encryption, brute-force lockout, audit log, HTTPS

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.

Portable Agent Skill for developing and auditing DeepSeek Harness plugins, with an optional profile-installable DSH bundle adapter.

Find and audit DeepSeek Harness (DSH) plugins inside the agent — live `dsh-plugin` topic search + source audit with trial-to-production install plans.

DSH plugin: triage port conflicts (reuse / switch / precise kill) — 端口占用处置

100% AI-native product delivery workflow plugin for DeepSeek Harness: full product-to-release pipeline (research → PRD → OpenSpec → parallel multi-agent → review loops → tests → release audit) with loop closure. 产品交付工作流插件：从产品到发布全流程，自带闭环，100% AI 原生，睡前启动醒来收货。

dsh-routines — scheduled agents for DSH: run a prompt on a cron, get the digest where you already are (file digests, chatnode delivery, unattended-safe)

Self-control guard plugin for DeepSeek Harness host exit and restart workflows.

Export DeepSeek Harness sessions to Markdown for review, blogging and audit. dsh plugin.

Security-audit skill pack for DeepSeek Harness (dsh): 5 agent skills - secret scan, dependency audit, supply-chain review, prompt-injection review, audit orchestration. Zero runtime code. Apache-2.0.

Fail-closed export-copy redaction for DeepSeek Harness session telemetry

Eight DeepSeek Harness plugins: persona, language guard, per-request vision fallback, python/windows write guards, cross-agent memory, image generation, and skill shell injection.

Advisory turn step-budget reminders for DeepSeek Harness — loop convergence guard (dsh-plugin)

Discover and install the best DeepSeek Harness plugins for a user's task

微软跨平台沙盒支持

Always-on vulnerability and breaking-change impact monitoring for DeepSeek Harness plugins.

agent-loop-workflow: 通用多 agent 协作工作流骨架 skill 插件 — Loop Guard/Handoff/Review→Close protocol

面向各类 AI 编程代理的保守代码修改守则 Skill，用于约束代理在已有项目中进行最小必要改动，避免无关重构，保护公共基础代码，并在支持深色模式的项目中优先使用动态颜色资源 An agent skill for keeping code changes small, scoped, and project-safe. Works with any AI coding tool that supports skills

A safety-first Git assistant for DeepSeek Harness that turns natural-language intent into validated Git commands with risk controls and optional execution.

Interactive terminal agent (fork of DeepSeek Harness): Claude Code-style REPL — full-screen UI, markdown rendering, session stats, permission cycling, custom OpenAI-compatible providers.

Independent-model advisory review for DeepSeek Harness: after each tool step, a reviewer model audits the agent's operations and injects concerns/guidance into the next step.

No repository or directory purpose description was collected.

DeepSeek Harness 自动审批门控：Flash 预判不可回补操作，安全自动批准、危险转人工（fail-safe）

模仿codex auto-review 的自动审批机制

DSH-native provider API configuration, model switching, and auditable token/cost analytics for DeepSeek Harness.

Inspect and compile Claude Code and Codex plugins into auditable DeepSeek Harness bundles.

Config drift detective for DSH: snapshots profile/plugin state, records changes.

Continual self-evolution plugin for DeepSeek Harness: versioned, auditable, rollback-safe harness state refined from session trajectories, with a benchmark-driven validation loop.

Difficulty-aware + taste-aware model routing plugin for DeepSeek Harness (DSH): hard/ambiguous tasks → strong model, UI/design → taste model, easy work stays cheap. Provider-agnostic, safe-by-default, calibrated on real bake-offs. MIT.

Local, zero-network, fail-closed secret preflight for DeepSeek Harness model requests.

DSH plugin: Git-backed project memory with line-addressable evidence, freshness tracking, and audit trail

DSH agent-fleet hygiene audit plugin: credential-file permissions, embedded git-remote credentials (masked), provider token literals. Read-only, zero-dependency, deterministic.

No repository or directory purpose description was collected.

Runtime tool policy, dangerous-command guard, and output redaction for DeepSeek Harness.

Plugin install, rollback, and audit as a service for DSH.

考研英语阅读命题助手，一个运行在 DeepSeek Harness（DSH）中的动态 Cordis 插件：自动抓取《卫报》《今日心理学》《经济学人》等外刊文章，调用大模型命制一套完整的考研英语（一）模拟卷，并保存到指定目录。|Postgraduate Entrance Exam English Reading Proposition Assistant, a dynamic Cordis plugin running in DeepSeek Harness (DSH): automatically crawls articles from foreign publications such as The Guardian, Psychology Today, etc

?? for DeepSeek Harness: first security plugin for dsh. Scans skills/MCP configs for prompt injection, homoglyphs, hidden Unicode, dangerous shell, credential leaks. DSH ????????

Runtime governance for DeepSeek Harness: progress governor, evidence freshness, completion guard, capability router (Same model. Same DSH. One runtime intervention.)

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

将DeepSeek Harness插件列表按来源分组展示——官方插件与自行安装插件，一目了然。

No repository or directory purpose description was collected.

DSH plugin auditor: pre-flight compatibility check for profile plugin combinations. DSH 插件审核器：安装新插件前扫描组合兼容性，预防启动崩溃。

dsh pm — discover, install, update, and manage dsh plugins from the CLI: multi-source search (awesome list + GitHub + npm), doctor audit, per-profile state

DeepSeek Harness plugin: run a live security health check on the deployment

Plugin Store for DeepSeek Harness (DSH): a graphical app-store inside the Harness Web UI — browse, search and one-click install GitHub dsh-plugins (topic:dsh-plugin / #dsh-plugin repos), with local ratings, dependency impact graphs, audit logging and a plugin scaffold guide.

No repository or directory purpose description was collected.

Auditable workspace-local project Wiki with a native Web UI for DeepSeek Harness

DSH / DeepSeek Harness plugin: runtime secret & PII redaction with mask, reversible vault and execution-time substitution

Crash-resilient remote execution world for DeepSeek Harness: ctx.fs/ctx.subprocess over an E2B sandbox with heartbeat keep-alive, transparent recovery, and workspace sync.

SSRF-resistant public-only HTTP(S) WebFetchProvider plugin for DeepSeek Harness

DSH compatibility plugin for redundant same-mode sandbox escalation requests

为 DeepSeek Harness 提供基于 Tailscale 的安全远程访问运营面板：一键健康检查、HTTPS 入口开关、macOS 代理绕过、中继服务器运维、ACL 生成。Tailscale remote-access control panel for DeepSeek Harness: health checks, HTTPS serve toggle, proxy bypass, relay ops, ACL helper.

Out-of-tree DeepSeek Harness guard that prevents stale todo_write state

Structured safe Git tools for DeepSeek Harness (dsh): git_status/diff/log/branch/stage/commit/stash/show + destructive-command guard

Virtual Workspaces for DeepSeek Harness: a dynamic Cordis Plugin that groups multiple project directories under one name for cross-project read/search/write, with native sidebar integration and sandbox-consistent escalation.

dsh-yolo-mode - an LLM-powered auto-approval plugin for DeepSeek Harness sandbox escalations (built-in presets + custom permission levels, fail-closed)

Long-task orchestration engine for AI agents: materialized state machine + T3 protocol dispatch + two-layer audit. Pure Python stdlib, schema-driven SKILL.

让测试绿了不等于你做对了 — 双轴收敛纪律 (spec-gaming orthogonal axis) 的 DeepSeek Harness 原生实现 | Two-axis convergence discipline for coding agents