Dasemu
b8f5d8253f
Replaces the slice-2 hardcoded installation with a full install/deploy pipeline: - SQLite persistence (better-sqlite3) with a single `installations` table and a draft -> cloning -> building -> built -> starting -> running/error/stopped state machine; env vars and bearer tokens are encrypted at rest (AES-256-GCM, key derived from MASTER_KEY). - Build/run orchestration: git clone (+ .env.example key detection), `nixpacks build` as a logged subprocess, dockerode container run/stop with forced host/port env vars and a TCP-connect health check before flipping an installation to "running". - REST API (installations CRUD, build/deploy/redeploy, logs, token regeneration) behind single-operator session-cookie auth with a basic login lockout. - Vite+React admin SPA: login, installation list, install wizard, and a detail page to configure env vars, trigger build/deploy/redeploy, view the build log, and copy the authenticated /mcp/<slug> URL. - Dockerfile (multi-stage, pulls in nixpacks + the Docker CLI) and docker-compose.yml for production deployment. Validated end-to-end against a real cloned repo: draft creation with env-key detection, build, the error path on a failed health check, and delete, all through the live API. Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>
remote-mcp-manager
Self-hosted manager for MCP servers: install one from a GitHub repo URL, configure
its env vars, build it with Nixpacks (no Dockerfile needed
in the source repo), run it in a container, and get back an authenticated
https://<your-domain>/mcp/<slug> URL guarded by a bearer token.
Quick start (production, Docker)
cp .env.example .env # fill in ADMIN_PASSWORD, SESSION_SECRET, MASTER_KEY, PUBLIC_BASE_URL
docker compose up -d --build
Put your own reverse proxy (nginx, etc.) in front of port 3000 for TLS and your public domain.
Local development (Podman)
Requires nixpacks and a docker CLI pointed at your Podman socket (e.g. via the
podman-docker package) on your PATH.
npm install
DOCKER_SOCKET_PATH=/run/user/$(id -u)/podman/podman.sock \
DATA_DIR=./data MASTER_KEY=... SESSION_SECRET=... ADMIN_PASSWORD=... \
npm run dev:server
Because rootless Podman isolates the host network namespace from the container bridge,
the manager itself must run as a container attached to mcp-net to reach the MCP
containers it manages — running it directly on the host only works for routes that
don't need to reach a sibling container (e.g. /health, /api/auth).
Known limitations
- Nixpacks needs a dependency manifest (
requirements.txt/pyproject.tomlfor Python,package.jsonfor Node) in the installed repo. Repos without one will fail to build; the full build log is always visible in the UI to diagnose this. - No per-container resource limits, image garbage collection, or multi-user support yet.
- Builds run one at a time.