diff --git a/README.md b/README.md
index cf34b20b..7a3b070e 100644
--- a/README.md
+++ b/README.md
@@ -2,190 +2,113 @@
# Pythinker Code
-### *Think first, then code. Your terminal-native AI engineering agent.*
-
-**An intelligent agent that reads your codebase, edits files, runs shell commands, searches the web, and iterates until the job is done.**
-**Powered by [Pythinker models](https://pythinker.com) — compatible with other LLM providers. All from the shell you already live in.**
-
-
-
-[](https://www.npmjs.com/package/@pymodel/pythinker-code)
-[](https://www.npmjs.com/package/@pymodel/pythinker-code)
-[](https://github.com/PyModel/pythinker-desktop-releases/releases)
-[](https://github.com/PyModel/pythinker-desktop-releases/releases/latest)
-[](https://github.com/PyModel/pythinker-desktop-releases/releases/latest)
-[](https://github.com/PyModel/pythinker-code/blob/main/package.json)
-[](https://github.com/PyModel/pythinker-code/blob/main/LICENSE)
-[](https://github.com/PyModel/pythinker-code/actions/workflows/ci.yml?query=branch%3Amain)
-
-[](https://www.typescriptlang.org/)
-[](https://agentclientprotocol.com/)
-[](https://modelcontextprotocol.io/)
-[](https://pymodel.github.io/pythinker-code/)
-[](https://code.pythinker.com)
-
-
-
-Website ·
-Quick Start ·
-Features ·
-IDE Integration ·
-MCP ·
-Architecture ·
+### A coding agent you can run as a desktop app
+
+[](https://www.npmjs.com/package/@pymodel/pythinker-code)
+[](https://code.pythinker.com/)
+[](LICENSE)
+[](https://code.pythinker.com/) | [](https://code.pythinker.com/)
+[](package.json)
+[](https://github.com/PyModel/pythinker-code)
+
+Download ·
+Capabilities ·
+Terminal ·
+Editor ·
Development
-
+
-
+
---
-## What is Pythinker?
-
-**Pythinker Code**is an open-source AI engineering agent that lives in your terminal. Give it a task — refactor a module, trace a bug, scaffold a feature — and it plans, executes tools, observes results, and keeps going until you are satisfied. It runs against **your repo, the shell, the web, and MCP tools**, with the model of your choice.
-
-It ships with first-class subagents for focused work — `coder` for scoped edits, `explore` for codebase reconnaissance, and `plan` for implementation design — all dispatched in parallel, isolated contexts from a single iterative loop.
-
-It speaks the [**Agent Client Protocol (ACP)**](https://agentclientprotocol.com/), so it slots cleanly into ACP-aware editors like Zed and JetBrains. It loads [**Model Context Protocol (MCP)**](https://modelcontextprotocol.io/) servers, so the same tools your other agents use just work. And it's hackable: subagents, skills, hooks, and plugins are all first-class extension points.
-
-> **Plan · Execute · Verify · Iterate.**One agent, one shell, one workflow. No tab-switching. No context loss. No magic.
-
----
-
-## Features
-
-
| - -### Terminal-First +## Get the desktop app -A purpose-built TUI tuned for long, focused agent sessions. Single-binary install, no Node.js required for end users, ready in milliseconds. +**[Download for macOS or Windows](https://code.pythinker.com/)** - | -
+Install it, open it, and describe a task. Pythinker then works on your project the way a colleague would: it reads the code, changes files, runs commands, checks what happened, and keeps going until the job is done.
-### Subagents & Skills
+The app runs the whole agent on your machine. It starts a local host bound to loopback, so nothing is exposed to your network. Closing the window hides the app to the tray instead of killing it, so a long session survives.
-Dispatch `coder`, `explore`, and `plan` subagents in parallel, isolated contexts. Load reusable repo-local instructions via `/skill: |
-
| +macOS gets a `.dmg`. Windows gets a per-user installer that does not ask for administrator rights. There is no Linux build yet, so on Linux use the terminal version or run `pythinker web` for the browser interface. -### ACP IDE Integration +More detail is in the [desktop guide](https://pymodel.github.io/pythinker-code/guides/desktop). -Run `pythinker acp` and any [Agent Client Protocol](https://agentclientprotocol.com/) editor — Zed, JetBrains, and more — gets a full Pythinker session inline. - - | -- -### MCP Tool Loading +--- -Add and authenticate MCP servers conversationally with `/mcp-config`. Plus a rich plugin ecosystem: skills, MCP servers, and data sources from the marketplace or GitHub. +## What the agent can do - | -
| +Everything below behaves the same in the app, in the terminal, and in your editor. -### Trust & Control +### Work on a real codebase -Approval flows to review tool calls before they run, a granular permission model, and lifecycle hooks to gate risky calls, audit decisions, and wire automation. +Pythinker searches and reads your repository, edits files, runs shell commands, and reads the output before it decides what to do next. It runs your tests, reads the failures, and tries again. Ask it to refactor a module, trace a bug, or fill in missing tests, and give it as much rope as you are comfortable with. - | -+### Split work across subagents -### Bring Your Own Model +Large tasks get delegated. A `coder` subagent makes scoped edits, `explore` maps unfamiliar parts of the repo, and `plan` designs the approach. They run in parallel with their own context, so the main conversation stays readable instead of filling with file dumps. -Works out of the box with [Pythinker models](https://pythinker.com); configurable for other compatible LLM APIs. +### Keep control of the tools - | -
|
+You see a tool call before it runs, and you approve it. The permission model lets you pre-approve the boring calls and hold the risky ones. Hooks fire on lifecycle events, so you can block a command, record a decision, or trigger something in your own systems.
-### Web & Dashboard UIs
+### Load your own tools and instructions
-A companion browser interface (`apps/pythinker-web`) and a session dashboard for replaying and debugging agent sessions (`apps/dashboard`).
+`/mcp-config` adds and authenticates [Model Context Protocol](https://modelcontextprotocol.io/) servers from inside a session, over stdio or HTTP, and remembers them for next time. Skills are repo-local instruction files that load on demand with `/skill: |
-+### Use the model you want -### Video Input +Pythinker models work out of the box. Other providers work by configuration, including any OpenAI-compatible endpoint and local ones. -Drop screen recordings into the conversation — let the agent *see* what is hard to describe. +### Show it instead of describing it - | -
+
@@ -212,164 +135,74 @@ Open a new conversation in Zed's Agent panel. For JetBrains setup, troubleshooti
---
-## MCP Tooling
-
-Pythinker loads [Model Context Protocol](https://modelcontextprotocol.io/) tools so the same servers your other agents use just work — stdio and HTTP transports, OAuth-backed servers, persistent config.
-
-Inside a session, `/mcp-config` manages everything conversationally:
-
-```
-/mcp-config # add, authenticate, test, and remove MCP servers
-```
-
-See the [configuration docs](https://pymodel.github.io/pythinker-code/configuration/config-files) for config-file based setup.
-
----
-
-## Architecture
-
-Pythinker Code is a **pnpm monorepo**. The CLI consumes capabilities through the SDK and never depends directly on internal engine packages.
-
-
-
-
+
+
| - -**▶ Run & iterate** -```sh -pnpm dev:cli # CLI in dev mode -pnpm dev:web # browser UI -pnpm dev:server # agent server -pnpm dashboard # session dashboard -``` - - | --**Verify** -```sh -pnpm test # test suite (Vitest) -pnpm typecheck # TypeScript check -pnpm lint # lint with oxlint -pnpm build # build all packages -``` - - | -
- Thanks for visiting Pythinker! -
+[code.pythinker.com](https://code.pythinker.com) · +[Download](https://code.pythinker.com/) · +[npm](https://www.npmjs.com/package/@pymodel/pythinker-code) · +[Docs](https://pymodel.github.io/pythinker-code/)