@zseta
基本情報
MCP-native workflow engine for your AI harness. Progi teaches your agent how you like to get things done. So you can do your best work without re-explaining your process or losing context between sessions.
以下の設定を使って、このサーバーを MCP 対応クライアントに追加してください。
{
"mcpServers": {
"progi": {
"command": "uvx",
"args": [
"progi"
]
}
}
}ツール一覧はまだありません
この掲載には表示できるツール一覧がまだありません。サーバーにツールがないという意味ではありません。対応機能はプロジェクトのドキュメントをご確認ください。
上のボタンで一時的なサンドボックスでこのサーバーを実行し、ツール一覧を取得できます。
Progi is an MCP-native workflow engine that teaches your AI agent your preferred processes. It stores structured workflows with per-step playbooks in a local SQLite database, survives context resets, and provides a monitoring web UI. It is intended for developers who want their agent to follow consistent, reusable procedures across sessions.
Describe your workflow in plain language to Progi. It runs a two-pass authoring process to create a structured skeleton and per-step playbooks, then persists everything with save_workflow. Run tasks with create_task, then start_or_continue_task to step through the workflow, and submit_output to advance. Monitor progress via the web UI (default at 127.0.0.1:8000) or disable it with PROGI_NO_WEB=1. Configure database path using the PROGI_DB_PATH environment variable.
Progi stores workflows as structured steps with playbooks in a persistent database. Unlike plain prompts, these workflows survive context resets, and each future task automatically picks up the latest playbook changes.
Set the `PROGI_DB_PATH` environment variable to an absolute path (defaults to the OS data directory). The web UI binds to `127.0.0.1:8000` by default; you can change the port with `PROGI_WEB_PORT` or disable it entirely with `PROGI_NO_WEB=1`.
All data is stored in a local SQLite database whose location is controlled by `PROGI_DB_PATH`. No external cloud or server is required.
Progi exposes an MCP‑compliant tool interface over stdio (when invoked via the `uvx` command) plus a web UI for monitoring. It does not provide its own network transport or authentication layer beyond the MCP protocol.
Pass 1 converts a plain-language description into a structured workflow skeleton using `get_process_skeleton_prompt`. Pass 2 authors each step’s detailed playbook using `get_playbook_authoring_prompt`. The result is persisted with `save_workflow`.
Model Context Protocol Servers
A search tool helps dev to solve the naming things problem.
Health by Dr Brad works out a health plan from your own results: which screening is due at your age, and evidence-based suggestions, each with the reason and the published guideline or study behind it. Your record is one
YC (S26) | Open Computer History | Continuously record your company computer work, map your workflows, help you find work worth automating, and power your agents' context
🚀 The fast, Pythonic way to build MCP servers and clients.
コメント