Search

Project-scoped CLI for SupaCloud users

DeepSeek Harness Web UI 工作台：聊天 / 编辑器（语法高亮） / 文件与 Git、侧边栏、模型工具

Validated event protocols shared by Doompi extensions

Synlo CLI — discover, approve, and synchronize your team's AI capabilities.

Language-neutral Node.js runtime capability registry.

Cordis plugin suite for omdsh terminal presentation, session interaction, commands, and human input

Right-hand widget rail for DeepSeek Harness Web UI: live session stats (turns, LLM/tool time, TTFT, speed, cache hit, tokens) plus OpenCode Go quota (rolling/weekly/monthly) proxied by a same-origin host route; an extensible widget registry as the base fo

Beta open-source SDK for streaming personalized motion experiences in web applications.

See what your coding agents actually did. Every session, measured locally.

Freebuff LLM router - a local OpenAI-compatible /v1 server that proxies free-mode Freebuff models (DeepSeek V4 Flash, MiMo, etc.) for any agent client.

Validation and deterministic packaging for OpenXiangda 2.0 AI skills.

Use a ChatGPT Codex subscription in DeepSeek Harness through standalone OAuth

Model-facing vision tool for DeepSeek Harness: analyze images via any OpenAI-compatible vision API.

No description available

Unified LLM API with automatic model discovery and provider configuration

DeepSeek Harness 外置视觉模型插件：inspect_image 把本地图片或 http(s) 图片 URL 发给任意 OpenAI 兼容端点，视觉模型看图的文字回答直接带回对话；附 Web UI 设置栏。

Mirendil-maintained models.dev catalog with explicit canonical base_model mappings

Drop-in token compression for LangChain.js agents — wrap your tools and The Lyceum dedupes repeated reads, caps runaway output, and strips machine noise before it burns your context window.

Claim XiaShe external Agent feeds and safely receive protected Skill packages.

Auto-detecting graph workflow with local checks, caching, selective retries, resume, and interactive reports