Search

Discover public APIs, search OpenAPI endpoints and schemas, preview requests, and generate type-safe SDK integrations with Pontx Hub

Install the UI Spec Codex skill and choose frontend components with exact bilingual terminology.

Framework-agnostic core engine for the IDmission Web SDK

Give text-only pi models vision — describe images with a vision model you pick via an interactive picker, then hand off the text description to non-vision models. Includes direct clipboard paste (image-paste integration).

Vision bundle + plugin for DeepSeek Harness: the describe_image tool reads local images and asks any configured multimodal route, with zero core changes

Local image understanding for the DeepSeek Harness: OCR, table-layout detection, and semantic description of textless images via macOS Vision, fully on-device (images never leave your Mac). 本地化识图，图片不出本机。

DSH（DeepSeek Harness）图片一体化插件：粘贴图片只进对话不进模型请求（llm/stream 清洗），并注册 see_image 工具让文本模型通过任意 OpenAI 兼容视觉 API 按需看图

MoonDesk — use ChatGPT Chat as a local coding agent, by Shattermoon.

Computer Use 插件：虚拟鼠标真人操作（screen_observe + computer_click 等 11 个模型友好工具，跨平台 cua-driver 引擎）

frappe client for multiplatform.one ecosystem

Type-safe {{variable}} templating for prompt files with fail-closed substitution

Fork of @narumitw/pi-goal with inline /goal invocation: autonomous single-objective /goal completion for Pi.

Interactive terminal UI for deepseek-harness, installed as a dsh profile bundle

Kapel — a multi-model orchestration coding agent for your terminal. The strongest model plans; cheaper or specialized workers execute, under a natural-language policy compiled to a deterministic runtime.

Pi extension for the Hetzner Experiments Inference API — OpenAI-compatible open-weight models, catalog aligned with models.dev

Nous Research Portal provider for oh-my-pi, with Portal OAuth and direct API-key authentication.

On-device speech recognition for Capacitor apps, via whisper.cpp

Text-to-speech (TTS) plugin for DeepSeek Harness — Fish Audio API only, bring your own key. Per-message read-aloud, auto-read toggle in the composer, settings for model / voice / encrypted API key / proxy. Third-party, not affiliated with Fish Audio.

dsh-voice-call — give the agent a voice it owns: the agent decides when to speak (offer_call), the human holds the answer key (接听/拒接/稍后). Local-first TTS via CrispASR + Qwen3-TTS CustomVoice, plain audio files under ~/.dsh/voice/. Fork of dsh-voice with a

Rewrite any AI draft so it reads as written by you — register-aware voice profiles built from your own writing, with deterministic quirk enforcement and a fidelity score.