Real-time cost tracking and budget management for Claude Code. Zero setup. No Docker. No Grafana. Just install and go.

A multi-agent application-security review harness for Claude Code (and, later, other AI agents). One router skill dispatches to a full offensive-security pipeline that maps a codebase, hunts vulnerabilities with a per-class knowledge base, chains findings into escalations, verifies real impact, and reports to README/JSON/SARIF/doc/PDF.

Kimi for Claude Code and Codex: code questions, reviews, fixes, and parallel tasks through your local Kimi Code CLI. Available through GitHub marketplaces and npm.

23 practical agent skills: TeamWorkflow, multi-model review, media ingestion, and writing. Original Claude Code packages plus a Codex desktop edition tuned for Astra. Source for clickai.dev.

Keep Claude Code’s prompt cache warm during breaks and show the estimated cost before a cold send.

让你通过浏览器安全地远程使用 DeepSeek Harness：登录、MFA 两步验证、角色权限和远程文件预览。 | Use DeepSeek Harness securely in a remote browser with login, MFA, roles, and remote file preview.

DSH Web 工作段折叠：保留原生一级，将正文之间的思考、工具、上下文合为一个二级，展开后显示原生三级。

DSH Web 壁纸引擎 Wallpaper Engine mpkg 背景插件：浏览器内直接解析 .mpkg（preview.gif 动态背景/内嵌 mp4 视频/多时段切换），整屏统一虚化/对话框/弹层/遮罩独立虚化、镜头缩放平移、时钟、冲突检测。

Audit and fix your Mac's security in one command. No agent, no signup, open source.

不把整台机器交出去，只送出挡路的决定和下一步：DeepSeek Harness 的审批、ask_user_question 提问与结果回复，以飞书卡片发到手机——桌面优先、只出站；执行过程实时可见，下一段任务一键即开。

Secure SSH, FTP/FTPS/SFTP file transfer, browser terminals, proxies, port forwarding, and session-scoped AI access for DeepSeek Harness

Use AI agents from the DSH session UI.

A persistent agent team for long-running collaboration in DeepSeek Harness

Local-first memory & context harness for coding agents

DSH 预设编辑器插件, 支持一键破甲.

DSH 原生 MUV 引擎 —— 正则脚本、状态栏级联（任意字段/行动选项）、变量追踪、LaTeX、宏展开、标签 CSS、visual 代码块渲染（信纸/终端/报纸/手机），内置 diag.mjs 诊断工具。需与伴生插件 dsh-muv-table 一起安装

Official-grade GitHub CI for DeepSeek Harness: composite action (review PRs / fix CI / report), polling PR review bot with idempotent inline comments, a status-check gate, and approval-gated PR/issue/repo/file tools

Merge every locally signed-in WorkBuddy account into DeepSeek Harness as one auto-failing-over model pool (multi-account rotation, live credits, daily check-in and model catalog).

Outline-Driven Development for Claude Code - 46 agents, 25+ skills, diagram-first methodology, AST-based editing, atomic commits.

Detects prompt injection by its effect on a sacrificial canary model, not just pattern matching: untrusted input hits a powerless model first, a behavioral check reads the residue, and it returns block, flag, or pass before your primary model acts. Inbound preflight sensor, not a guarantee.

Browser automation clicks buttons. OpenTabs calls APIs.

Long-lived AI companions with WeChat, Matrix and Mattermost channels for DeepSeek Harness

DSH permission-gate for the DeepSeek Harness 0.1.5 line: a single self-sufficient, deterministic-first, fail-closed gate covering P0 hard-deny -> P1 session grant -> P2 static rule (allow/deny) chain -> P3 optional LLM semantic classifier -> P4 ask, with

Automatic model catalog synchronization for API-key DeepSeek Harness providers.