Reusable AI skills and capabilities for agent workflows
mangoc
Sync one sub2api instance's groups, keys, and models into dsh-llm-pi-ai provider routes: panel-driven login (password / OAuth), one relay key one provider.
...moreGitHub Actions
Authoring and distribution for agent skills
ohmycodeapp
Browse every public Agent Skill grouped by theme, multi-select and install them through `npx skills add`.
hirenf14
Cloud Agent Fleet Worker — machine companion and operator CLI.
matyik
Computer-use for DeepSeek Harness on macOS: let an agent see the screen, drive the mouse and keyboard, and open apps and URLs — no helper binary, permissions come from the Harness app itself.
...morebemlerlabs
Local-first AI assistant — one-command install wrapper (delegates to install.sh).
ferrisching
输入框提示词联想插件：助手回复结束后预测你的下一句话，以灰色 ghost text 浮在输入框里，Tab 采纳、Esc 忽略 ｜ Codex-style composer prompt predictions for DeepSeek Harness: your next message ghosted into the composer, accepted with Tab.
...morexiaohai78
Mode manager for DeepSeek Harness (dsh): switch what each agent mode can do, edit its system prompt and tool descriptions, manage skills per mode with versions and imports from Codex and Claude Code, create modes from templates, and merge built-in updates
...moreGitHub Actions
Local-first semantic search for your documents, PDFs, images, audio and video.
GitHub Actions
Verify you are talking to LibertAI inference running in an attested confidential VM, then talk to it directly.
GitHub Actions
On-device text-to-speech in the browser: Phonon compiled to WebAssembly, streaming 24 kHz speech from a worker.
GitHub Actions
Text-to-speech for the browser and Node with pluggable engines — starting with the free Edge read-aloud service.
navanatech
Bodhi (Navana Tech) Speech-to-Text and Text-to-Speech SDK for Node.js
germey
Turn text into Fish Audio speech, use reference voices, discover voice models, and track synthesis tasks in n8n with AceDataCloud.
...moreGitHub Actions
Indra OCR running locally in browser WebAssembly workers
ryan_alpaca
Lightweight ONNX captcha OCR model (4-char alphanumeric), cropped from ddddocr. 5MB, runs in browser via onnxruntime-web.
...moreGitHub Actions
Coordinate one change across many repos with GitHub Copilot CLI.
expiracy
Transparent, durable multi-agent workflows for OpenCode 2
crunaldo
Safe project-scoped multi-agent autopilot for Codex with interactive setup, diff, doctor, and reusable Goal contracts
syn7xx
Deterministic multi-agent Workflows for OpenCode: TypeScript scripts that fan work out to subagents, with typed results, a live TUI panel, a web app and a versioned protocol.
...moreshirtiny
A Pi extension that displays context-window usage and message composition
shifengintelli
把 Claude Code、Codex 等 AI 编码智能体每个会话窗口的状态，实时同步到桌面硬件屏幕
cwsong
DSH 生图插件：在设置中配置多个生图供应商/模型，并在对话中通过 image_generate 调用当前选中的模型。
cwsong
DSH 生图插件：在设置中配置多个生图供应商/模型，并在对话中通过 image_generate 调用当前选中的模型。