pi-ai-backed DeepSeek adapter for the DeepSeek Harness LLM seam (design-verification twin of dsh-llm-deepseek)

Append-only auto-sync for dsh llm-pi-ai provider catalogs (models.dev + OpenAI /models) with backup and restore. Discovers providers from your settings — no hard-coded provider in JSON.

Auto-complete missing model display name, context window, max output tokens, and reasoning efforts in the llm-pi-ai settings namespace.

pi-ai-backed DeepSeek adapter for the Tianshu Harness LLM seam (design-verification twin of dsh-llm-deepseek)

Background auto-fill of llm-pi-ai model capability fields from overrides → pi-ai catalogs → models.dev

Model Pro — DSH plugin for llm-pi-ai provider lifecycle: create/edit, custom headers, enable/disable with uninstall-restore safety, smart routing, composites, observability, and per-model connectivity tests

DeepSeek chat-completions adapter for the DeepSeek Harness LLM seam

pi-ai-backed DeepSeek adapter for the DeepSeek Harness LLM seam (design-verification twin of dsh-llm-deepseek)

General-purpose agent with transport abstraction, state management, and attachment support

Unified LLM API with automatic model discovery and provider configuration

Local SQLite memory engine for Oh My Pi agents

Models settings and official-DeepSeek first-run routing over one live provider/settings/credential join

Settings page to declare per-model reasoningEfforts and openai-completions compat on hand-declared llm-pi-ai routes. Writes the official namespace; does not intercept llm/stream or replace the Models page.

General-purpose agent with transport abstraction, state management, and attachment support

Unified LLM API with automatic model discovery and provider configuration

DSH Web GUI 模型能力配置：在「设置 → 模型」的模型目录编辑器内，为每个已配置的模型直接设置是否支持图像与思考等级（reasoning efforts），写入 llm-pi-ai 设置命名空间。

Harness-backed AI testing on top of Vitest.

General-purpose agent with transport abstraction, state management, and attachment support

Detect agentic coding environments and AI assistant tools

Bitmap-frame context compression for vision-capable LLMs