15
Published Tools
4
Total Stars
3,720
Weekly Downloads
0
GitHub Followers
17
Public Repos
100/100
Avg Security
Published Tools
1 MCP Server13 Skills1 Agentacross 7 categoriespaulrichez
Charts for aparté via Chart.js — a Chart.js config a model writes, or a tool returns, drawn as it streams.
paulrichez
Render an AG-UI agent in aparté — Agno, LangGraph, Mastra, CrewAI, Pydantic AI, TanStack AI: the agent runs on its server, aparté draws the chat, its tool calls and its approvals.
...morepaulrichez
Charts for aparté via Apache ECharts — an ECharts option a model writes, or a tool returns, drawn as it streams.
paulrichez
The aparté documentation as an MCP server — a coding agent searches the guides, reads a page or a whole topic (tools & approval, bring your own loop, theming & the UI kit…), always the version that ships, straight from apartejs.dev.
...morepaulrichez
Run LLMs 100% in the browser via Transformers.js (WebGPU/WASM) — a local, keyless AI provider for aparté. Streams tokens off the main thread in a Web Worker.
...morepaulrichez
One OpenAI-compatible chat-completions adapter for any /v1 endpoint — OpenAI, Mistral, OpenRouter, Z.ai, Groq, LM Studio, Ollama and friends, via presets or plain config.
...morepaulrichez
Put aparté on an n8n chat workflow — the Chat Trigger's agent runs in n8n, with its memory and its tools; aparté draws the chat, streamed or not.
...morepaulrichez
Bridge any Vercel AI SDK model into aparté — bring your own @ai-sdk/* package (Anthropic, Google, OpenAI, 25+ vendors) and aparté renders it.
...morepaulrichez
The ask_user tool for aparté — the AI asks the user a structured choice, presented via the core elicitation panel.
paulrichez
Provider + model selector web component (<aparte-model-selector>) for aparté — grouped dropdown, BYOK.
paulrichez
Approval modes for aparté tool calls — plan, ask, auto-edit, auto — from a classification of your tool names, with a switch for the composer's toolbar.
...morepaulrichez
The agent loop on its own: runStreamAgent streams a model's turns and runs its tools, in the browser or in Node — zero runtime dependencies.
...morepaulrichez
Conversation compaction for aparté — summarise the turns that no longer fit the model's window, keep the recent ones verbatim, and answer the context gauge's aparte-compact.
...morepaulrichez
AI chat and agent UI components as Web Components, with the agent loop inside — zero third-party dependencies
paulrichez
Artifacts for aparté — a create_artifact tool the model calls, a rich Code/Preview card that renders its result, and the <artifact> tag grammar for a model that writes one in its prose.
...more