{"id":51510298,"url":"https://github.com/man4ish/omnibioai-launcher","last_synced_at":"2026-07-08T05:01:59.414Z","repository":{"id":361315508,"uuid":"1254005146","full_name":"man4ish/omnibioai-launcher","owner":"man4ish","description":"Standalone React app that opens OmniBioAI registry objects directly in JupyterLab, VS Code, or RStudio — automatically injects object ID, API base URL, and auth token as environment context so researchers can start analyzing immediately without manual setup.","archived":false,"fork":false,"pushed_at":"2026-06-08T06:47:15.000Z","size":210,"stargazers_count":0,"open_issues_count":0,"forks_count":0,"subscribers_count":0,"default_branch":"main","last_synced_at":"2026-06-08T08:05:43.400Z","etag":null,"topics":["bioinformatics","javascript","jupyter-notebook","omnibioai","react","rstudio","visual-studio-code"],"latest_commit_sha":null,"homepage":"","language":"JavaScript","has_issues":true,"has_wiki":null,"has_pages":null,"mirror_url":null,"source_name":null,"license":null,"status":null,"scm":"git","pull_requests_enabled":true,"icon_url":"https://github.com/man4ish.png","metadata":{"files":{"readme":"README.md","changelog":null,"contributing":null,"funding":null,"license":null,"code_of_conduct":null,"threat_model":null,"audit":null,"citation":null,"codeowners":null,"security":null,"support":null,"governance":null,"roadmap":null,"authors":null,"dei":null,"publiccode":null,"codemeta":null,"zenodo":null,"notice":null,"maintainers":null,"copyright":null,"agents":null,"dco":null,"cla":null}},"created_at":"2026-05-30T03:14:17.000Z","updated_at":"2026-06-08T06:47:19.000Z","dependencies_parsed_at":null,"dependency_job_id":null,"html_url":"https://github.com/man4ish/omnibioai-launcher","commit_stats":null,"previous_names":["man4ish/omnibioai-launcher"],"tags_count":0,"template":false,"template_full_name":null,"purl":"pkg:github/man4ish/omnibioai-launcher","repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/man4ish%2Fomnibioai-launcher","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/man4ish%2Fomnibioai-launcher/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/man4ish%2Fomnibioai-launcher/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/man4ish%2Fomnibioai-launcher/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/man4ish","download_url":"https://codeload.github.com/man4ish/omnibioai-launcher/tar.gz/refs/heads/main","sbom_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/man4ish%2Fomnibioai-launcher/sbom","scorecard":null,"host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":286080680,"owners_count":35252324,"icon_url":"https://github.com/github.png","version":null,"created_at":"2022-05-30T11:31:42.601Z","updated_at":"2026-05-26T15:22:16.424Z","status":"online","status_checked_at":"2026-07-08T02:00:06.796Z","response_time":61,"last_error":null,"robots_txt_status":"success","robots_txt_updated_at":"2025-07-24T06:49:26.215Z","robots_txt_url":"https://github.com/robots.txt","online":true,"can_crawl_api":true,"host_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub","repositories_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories","repository_names_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repository_names","owners_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners"}},"keywords":["bioinformatics","javascript","jupyter-notebook","omnibioai","react","rstudio","visual-studio-code"],"created_at":"2026-07-08T05:01:57.795Z","updated_at":"2026-07-08T05:01:59.361Z","avatar_url":"https://github.com/man4ish.png","language":"JavaScript","funding_links":[],"categories":[],"sub_categories":[],"readme":"# OmniBioAI Launcher\n\nA browser-based gateway to interactive analysis environments for the OmniBioAI platform.\nThe launcher operates in two independent modes: opening a specific registry object in your\npreferred IDE, and starting/stopping long-running IDE services backed by Docker containers.\n\nThis repository is intentionally separate from\n[omnibioai-sdk](https://github.com/man4ish/omnibioai_sdk), the pure Python API client.\nThe launcher is the browser entry point; the SDK is for programmatic use inside notebooks\nand scripts.\n\n---\n\n## Overview\n\n| Mode | What it does |\n|---|---|\n| **Object Launch** | Browse the registry, select an object, open it in JupyterLab, VS Code, or RStudio with context pre-loaded |\n| **IDE Services** | Start / stop containerised JupyterLab, RStudio, and VS Code Server from the Launcher UI |\n\nThe two modes are independent — IDE Services can be used without an object context, and\nObject Launch works with any running JupyterLab instance.\n\n---\n\n## Supported Environments\n\n| Environment | Description | Default port |\n|---|---|---|\n| **JupyterLab** | Full bioinformatics kernel (scanpy, DESeq2, scVelo, cellxgene …) | 8888 |\n| **RStudio** | R with Bioconductor — Seurat, DESeq2, scran, monocle3, tidyverse | 8787 |\n| **VS Code Server** | Python + R + Nextflow + WDL extensions, all packages from above | 8080 |\n\n---\n\n## Running in OmniBioAI Stack (recommended)\n\nThe Launcher is managed automatically by OmniBioAI Studio.\nNo manual startup required — it starts with the full stack:\n\n```bash\ncd ~/Desktop/machine/omnibioai-studio\ndocker compose up -d launcher\n```\n\nAccess at: `http://localhost/_svc/sdk` (via nginx, JWT required)\nDirect access (localhost only): `http://localhost:5190`\n\nThe Launcher backend API runs on port 3001 internally.\nIDE container lifecycle (start/stop/status) is handled via\nthe Docker socket — no additional configuration needed.\n\n---\n\n## Quick Start — IDE Services\n\nThe fastest way to get all three environments running locally:\n\n```bash\n# Clone and start\ngit clone https://github.com/man4ish/omnibioai-launcher.git\ncd omnibioai-launcher\ndocker-compose up -d\n```\n\n| Service | URL | Default credential |\n|---|---|---|\n| JupyterLab     | http://localhost:8888 | token: `$JUPYTER_TOKEN` (set in .env)    |\n| RStudio        | http://localhost:8787 | password: `$RSTUDIO_PASSWORD` (set in .env) |\n| VS Code Server | http://localhost:8080 | password: `$VSCODE_PASSWORD` (set in .env)  |\n\nStop all services:\n\n```bash\ndocker-compose down\n```\n\nOverride credentials or data paths with environment variables:\n\n```bash\nJUPYTER_TOKEN=mysecret OMNIBIOAI_DATA_DIR=/data/myproject docker-compose up -d\n```\n\n---\n\n## Quick Start — Object Launch\n\nThe Launcher UI is a React single-page app served on port 5190.\n\n**With a running backend:**\n\n```bash\ncp .env.example .env    # fill in REACT_APP_OMNIBIOAI_BASE_URL and token\nnpm install\nnpm start               # dev server at http://localhost:3000\n```\n\n**Via Docker:**\n\n```bash\ndocker build -t omnibioai-launcher .\ndocker run -p 5190:5190 omnibioai-launcher\n```\n\n**Direct link from any page:**\n\n```html\n\u003ca href=\"http://127.0.0.1:5190/?object_id=56d3fc3a-709b-4ed0-bf17-8cb73c6746b0\"\u003eAnalyze\u003c/a\u003e\n```\n\nIf no `object_id` is given, the app opens a searchable registry list. Selecting an object\nshows a detail view (metadata, lineage, job log) and a button to open it in an environment.\n\n---\n\n## Pre-installed Packages\n\n### JupyterLab (`docker/jupyter/Dockerfile`)\n\nBase image: `jupyter/datascience-notebook:latest`\n\n**Python** — scanpy, anndata, scVelo, squidpy, pyDEA, gseapy, biopython, pysam,\ncellxgene, leidenalg, harmonypy, decoupler, pydeseq2, omnipath\n\n**R / Bioconductor (via conda)** — DESeq2, edgeR, limma, Seurat\n\n### RStudio (`docker/rstudio/Dockerfile`)\n\nBase image: `rocker/rstudio:4.3.2`\n\n**Bioconductor** — DESeq2, edgeR, limma, Seurat, clusterProfiler, EnhancedVolcano,\nComplexHeatmap, SingleCellExperiment, scran, scater, monocle3\n\n**CRAN** — tidyverse, ggplot2, pheatmap, RColorBrewer, patchwork, cowplot\n\n### VS Code Server (`docker/vscode/Dockerfile`)\n\nBase image: `codercom/code-server:latest`\n\n**Extensions** — ms-python.python, REditorSupport.r, nextflow-io.nf-lang, broadinstitute.wdl\n\n**Python packages** — scanpy, anndata, scVelo, pydeseq2, gseapy, biopython, pysam\n\n---\n\n## Architecture\n\n```\nOmniBioAI Studio\n      |\nLauncher UI  (React, port 5190)\n      |\n  ┌───┴──────────────────────────┐\n  │  Object Launch               │  IDE Services\n  │  (registry object context)   │  (container lifecycle)\n  └───┬──────────────────────────┘\n      |                                  |\n  Open object in:               docker-compose up/down\n  - JupyterLab  (URL + token)   GET  /api/launcher/status/{tool}\n  - VS Code     (env var copy)  POST /api/launcher/start/{tool}\n  - RStudio     (.R download)   POST /api/launcher/stop/{tool}\n```\n\nThe `IdeCard` component in the Launcher UI polls `GET /api/launcher/status/{tool}` every\n5 seconds. Clicking **Launch** calls `POST /api/launcher/start/{tool}`, polls until the\ncontainer reports `running`, then opens the service URL in a new tab. A **Stop** button\nappears while the container is running.\n\n### OmniBioAI nginx routing\n\nIn production the Launcher is accessed via nginx:\n\n```\nhttp://localhost/_svc/sdk  →  launcher:5190  (JWT required)\n```\n\nThe `/api/launcher/*` endpoints are proxied to the Express backend\non port 3001 inside the container.\n\n---\n\n## API Endpoints\n\n### Object registry (existing)\n\n| Method | Endpoint | Purpose |\n|---|---|---|\n| `GET` | `/api/dev/objects/` | Paginated object list (`search`, `type` filters) |\n| `GET` | `/api/dev/objects/{id}/` | Single object detail |\n| `GET` | `/api/dev/objects/?parent_id={id}` | Children / siblings for lineage view |\n| `POST` | `/api/dev/launch/rstudio/` | Signal backend to open RStudio after R script download |\n\n### IDE services (new)\n\n| Method | Endpoint | Purpose |\n|---|---|---|\n| `GET` | `/api/launcher/status/{tool}` | Container status (`running` / `starting` / `stopped`) |\n| `POST` | `/api/launcher/start/{tool}` | Start the IDE container |\n| `POST` | `/api/launcher/stop/{tool}` | Stop the IDE container |\n\n`{tool}` is one of `jupyter`, `rstudio`, `vscode`.\n\nAll requests carry `Authorization: Bearer \u003ctoken\u003e`.\n\n---\n\n## Docker Images\n\nPre-built images are published to the GitHub Container Registry:\n\n```\nghcr.io/man4ish/omnibioai-jupyter:1.0\nghcr.io/man4ish/omnibioai-rstudio:1.0\nghcr.io/man4ish/omnibioai-vscode:1.0\n```\n\nTo rebuild and push:\n\n```bash\nexport CR_PAT=$(gh auth token)\necho $CR_PAT | docker login ghcr.io -u man4ish --password-stdin\n\nfor tool in jupyter rstudio vscode; do\n  docker build \\\n    -t ghcr.io/man4ish/omnibioai-${tool}:1.0 \\\n    -f docker/${tool}/Dockerfile docker/${tool}/\n  docker push ghcr.io/man4ish/omnibioai-${tool}:1.0\ndone\n```\n\n---\n\n## Environment Variables\n\n### Launcher UI (baked in at build time, prefixed `REACT_APP_`)\n\n| Variable | Default | Purpose |\n|---|---|---|\n| `REACT_APP_OMNIBIOAI_BASE_URL` | `http://127.0.0.1:8000` | OmniBioAI backend API base URL |\n| `REACT_APP_OMNIBIOAI_TOKEN` | `dev` | Bearer token for all API requests |\n| `REACT_APP_JUPYTER_BASE` | `http://127.0.0.1:8890` | JupyterLab host for object-launch URL |\n| `REACT_APP_JUPYTER_TOKEN` | `devtoken` | JupyterLab auth token (`?token=`) |\n| `REACT_APP_USE_MOCK` | `false` | Use hardcoded mock data without a backend |\n\n### docker-compose services (runtime)\n\n| Variable | Default | Purpose |\n|---|---|---|\n| `JUPYTER_TOKEN` | `omnibioai` | JupyterLab authentication token |\n| `RSTUDIO_PASSWORD` | `omnibioai` | RStudio login password |\n| `VSCODE_PASSWORD` | `omnibioai` | VS Code Server login password |\n| `OMNIBIOAI_DATA_DIR` | `./data` | Host path mounted as `/data` in all containers |\n| `OMNIBIOAI_WORK_DIR` | `./work` | Host path mounted as `/work` in all containers |\n\n\u003e **Security note:** `JUPYTER_TOKEN`, `RSTUDIO_PASSWORD`, and\n\u003e `VSCODE_PASSWORD` default to `omnibioai`. Change these in\n\u003e `omnibioai-studio/.env` before production use.\n\n---\n\n## Development\n\n```bash\ncp .env.example .env    # edit variables for your local setup\nnpm install\nnpm start               # dev server on http://localhost:3000\n```\n\nThe `proxy` field in `package.json` forwards `/api/*` calls to `http://127.0.0.1:8000`,\nso the OmniBioAI backend must be running locally during development.\n\n**Production build:**\n\n```bash\nnpm run build\n# serve the build/ output with any static file server\nnpx serve -s build -l 5190\n```\n\n**Launcher Docker image** (nginx, port 5190):\n\n```bash\ndocker build -t omnibioai-launcher .\n\n# Override backend at build time\ndocker build \\\n  --build-arg REACT_APP_OMNIBIOAI_BASE_URL=https://api.omnibioai.com \\\n  --build-arg REACT_APP_OMNIBIOAI_TOKEN=mytoken \\\n  -t omnibioai-launcher .\n\ndocker run -p 5190:5190 omnibioai-launcher\n```\n\n---\n\n## Mock Mode\n\nSet `REACT_APP_USE_MOCK=true` (or pass `?object_id=test` in the URL) to run entirely on\nhardcoded data without a backend. Useful for UI development and screenshots.\n\n---\n\n## Project Structure\n\n```\nomnibioai-launcher/\n├── docker/\n│   ├── jupyter/\n│   │   └── Dockerfile          — JupyterLab + bioinformatics packages\n│   ├── rstudio/\n│   │   └── Dockerfile          — RStudio + Bioconductor stack\n│   └── vscode/\n│       └── Dockerfile          — VS Code Server + Python/R/workflow extensions\n├── src/\n│   ├── App.jsx                 — View logic: list, detail, launcher\n│   ├── App.css                 — Dark-theme styles\n│   ├── index.js                — React root mount\n│   └── components/\n│       ├── EnvCard.jsx         — Clickable environment tile (object launch)\n│       ├── IdeCard.jsx         — IDE service card with status polling\n│       ├── ObjectCard.jsx      — Object metadata display\n│       ├── InstallModal.jsx    — Fallback modal when desktop app not found\n│       └── Toast.jsx           — Ephemeral status notification\n├── public/\n│   └── index.html\n├── docker-compose.yml          — IDE services orchestration\n├── .env.example\n├── package.json\n├── nginx.conf\n└── Dockerfile                  — Launcher UI (React → nginx)\n```\n\n---\n\n## Related Services\n\n| Service | Role |\n|---------|------|\n| `omnibioai-studio` | Manages Launcher container lifecycle |\n| `omnibioai` | Workbench backend — object registry API |\n| `omnibioai-api-gateway` | JWT enforcement on `/_svc/sdk` |\n| `omnibioai-control-center` | Health monitoring (launcher:5190) |\n| `omnibioai-sdk` | Python SDK client — programmatic alternative to Launcher UI |\n\n---\n\n## License\n\nApache License 2.0\n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fman4ish%2Fomnibioai-launcher","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Fman4ish%2Fomnibioai-launcher","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fman4ish%2Fomnibioai-launcher/lists"}