by Hive-Academy
Enables AI‑driven development across desktop, VS Code, and CLI by indexing codebases, persisting hybrid memory, and orchestrating specialized agents to automate coding, testing, and deployment tasks.
Ptah provides a provider‑agnostic AI development environment that analyses a repository, builds a searchable map of its architecture, and runs a team of context‑aware agents to perform code generation, review, testing, and automation. The experience is available as a native Electron desktop app, a VS Code sidebar/webview extension, and a headless CLI.
npx @hive-academy/ptah-cli init --human # interactive setup
npx @hive-academy/ptah-cli doctor # verify installation
npx @hive-academy/ptah-cli session start --task "explain this repo"
The CLI works in CI pipelines, scripts, or any headless environment.
ptah_* tool catalog (workspace analysis, LSP diagnostics, AST queries, symbol search, dependency graphs, browser automation, sandboxed code execution).Q: Do I need an internet connection? A: Only when using remote model providers. Local models (Ollama) run fully offline.
Q: How are API keys stored? A: Keys are saved in the OS secure store (Keychain, Credential Manager, or secret‑service) and never leave the machine.
Q: Can I mix different model vendors in one session? A: Yes. Each sub‑agent can be assigned its own provider, allowing a security‑focused model to review code written by a faster model.
Q: Is there a way to pause or cancel an agent run? A: Agents expose tool calls in an execution tree; you can approve or reject each call from any connected gateway before it executes.
Q: How do I add my own plugins or skills?
A: Use the plugin manager in the desktop app or VS Code extension, or drop a skill JSON file into the .ptah/skills/ folder; the system will validate hashes before loading.
An AI dev team that learns your architecture and keeps shipping.
Persistent memory, a staffed team of agents, and always-on delivery — in a desktop app, a VS Code extension, and a headless CLI.
Website · Download · Documentation · VS Code extension
The video above is a real session, not a mockup. It shows memory recall, sub-agent fan-out, and scheduled runs.
Short feature clips:
| Clip | Topic |
|---|---|
| Install | Install the desktop app on your platform |
| Sign in | Accounts, keys, and credential storage |
| Workspace analysis | Scan a repository and detect its stack |
| Agent generation | Generate a project-specific agent team |
| Agent orchestration | Spawn, monitor, and steer headless agents |
| Built-in MCP server | The ptah_* tool catalog and code execution |
| Model providers | Add providers, keys, and models |
| Plugins | Install skills and workflow packs |
Ptah is an AI development environment for real projects. It indexes your repository before the first message, keeps the decisions it makes after the last one, and runs a team of specialized agents against that shared map.
One core powers three surfaces:
| Surface | What it is | Best for |
|---|---|---|
| Desktop app | Native app for Windows, macOS, and Linux | The complete product, including always-on runs |
| VS Code extension | Sidebar and full-panel webview | Work that stays inside the editor |
| CLI | Headless JSON-RPC 2.0 over stdio | CI pipelines, scripts, and agent-to-agent use |

Ptah builds a map of your codebase and keeps it.
Read more: Memory · Workspace analysis

Read more: Agents · Skill synthesis · Tribunal
Read more: Automation

ptah_* tool catalog: workspace analysis, LSP diagnostics, AST analysis, symbol search, dependency graphs, browser automation, agent spawning, and sandboxed code execution.ptah mcp-serve) — drive Ptah from any MCP-compliant host. External orchestrators can spawn, monitor, and steer agents, or delegate a whole task..ptah/specs/.
Ptah is provider-agnostic. Bring the vendors you already use:
Keys stay on your machine, in the operating system secure store. Mix vendors so the model that reviews a security-sensitive diff is not the model that wrote it.
Read more: Providers
npm i -g @hive-academy/ptah-cli
ptah init --human # guided setup on a terminal
ptah doctor # confirm the install is ready
ptah session start --task "explain this repo"
The CLI runs standalone. It does not need VS Code or the desktop app.
Read more: Getting started · CLI guide
Ptah is free and open source under the MIT license. There is no trial and no credit card.
Ptah Builders is a paid membership for cohort sessions, private support, and community packs. The product itself stays free. Join the waitlist at ptah.live.
Ptah is an Nx monorepo. The apps and libraries share one hexagonal core.
apps/ # desktop app, VS Code extension, webview, CLI, TUI,
# license server, landing page, docs site, e2e suites
libs/backend/ # runtime-agnostic services; ports live in platform-core,
# adapters in platform-vscode | platform-electron | platform-cli
libs/frontend/ # Angular 21 feature libraries (signals, OnPush)
libs/shared/ # the one bridge between frontend and backend
libs/api/ # NestJS libraries for the license server
libs/web/ # Angular libraries for the marketing site
git clone https://github.com/Hive-Academy/ptah-extension.git
cd ptah-extension
cp .env.example .env
npm install
npm run dev # watch the extension and the webview
npm run electron:serve # run the desktop app
npm run cli:dev # run the headless CLI
npm run docs # serve the documentation site
npm run landing # serve the marketing site
npm run lint:all
npm run typecheck:all
npm run test
npx nx graph # visualize the dependency graph
Architecture rules, coding standards, and per-project guides are in the instruction files at the repository root and in each app and library folder.
Read CONTRIBUTING.md before you open a pull request. Report a security issue through SECURITY.md.
MIT © Ptah Orchestra
Please log in to share your review and rating for this MCP.
Explore related MCPs that share similar capabilities and solve comparable challenges
by headroomlabs-ai
Compress tool outputs, logs, files, RAG chunks, and conversation history before they reach the LLM, keeping answers identical while saving up to 95% of tokens for JSON payloads.
by modelcontextprotocol
A Model Context Protocol server for Git repository interaction and automation.
by zed-industries
A high‑performance, multiplayer code editor designed for speed and collaboration.
by modelcontextprotocol
Model Context Protocol Servers
by modelcontextprotocol
A Model Context Protocol server that provides time and timezone conversion capabilities.
by cline
An autonomous coding assistant that can create and edit files, execute terminal commands, and interact with a browser directly from your IDE, operating step‑by‑step with explicit user permission.
by upstash
Provides up-to-date, version‑specific library documentation and code examples directly inside LLM prompts, eliminating outdated information and hallucinated APIs.
by daytonaio
Provides a secure, elastic infrastructure that creates isolated sandboxes for running AI‑generated code with sub‑90 ms startup, unlimited persistence, and OCI/Docker compatibility.
by continuedev
Enables faster shipping of code by integrating continuous AI agents across IDEs, terminals, and CI pipelines, offering chat, edit, autocomplete, and customizable agent workflows.
{
"mcpServers": {
"ptah-mcp": {
"command": "npx",
"args": [
"@hive-academy/ptah-cli",
"mcp-serve"
],
"env": {}
}
}
}claude mcp add ptah-mcp npx @hive-academy/ptah-cli mcp-serve