Why VS Code en

Why We Now Recommend Wisp Science

Update (2025): This page originally recommended VS Code because its AI chat interface (e.g., Copilot Chat) lets beginners ask AI while coding. A year later, Wisp Science takes this idea to its logical conclusion — it's not just an editor with an AI sidebar, but an AI research workbench built specifically for scientific computing. We now recommend Wisp Science for running FigureYa.

---

From VS Code to Wisp Science

The original case for VS Code over RStudio was: AI assistance means beginners are no longer stuck on syntax and configuration. Wisp Science goes further:

Original reason to choose VS Code	How Wisp Science evolves it
Integrated chat panel	AI is native — not just chat, but an agent that reads/writes files, runs code, and manages projects
Multi-language support (R, Python, etc.)	Built-in persistent Python and R kernels; variables persist across cells and turns
Required manual extension installs & config	Works out of the box: bundled MCP bioinformatics tools (PubMed, GEO, ~80 databases), SSH/WSL/GPU remote compute
AI could only answer questions	AI can execute full analysis: retrieve data → write code → generate figures → record decisions

---

Key Advantages of Wisp Science

1. AI-Native Research Workbench

• Bring your own model: Supports OpenAI-compatible and Anthropic models, or drive Codex / Claude Code over ACP.
• Beyond chat: The agent reads and writes project files, runs shell, loads reusable Skills (SKILL.md). Approval gates stay on by default.
• Project memory: Restart and the full history is back. Undo a turn's file edits. Use @ for files, # for saved sessions, / for skills.

2. Persistent Python/R Kernels

• Each conversation gets its own isolated kernel; variables persist across cells and turns.
• Parallel conversations never share state.

3. Remote Compute

• Register local, WSL, and SSH hosts once.
• Probe hardware, submit long Runs with live logs.

4. Built for Science

• ~80 scientific databases (PubMed, GEO, etc.) via bundled MCP servers.
• Offline previews for notebooks, PDFs, Office files, and images.
• Isolated explorations to try a direction without touching the mainline.
• Publication Workspace that freezes manuscript revisions into verifiable Evidence Capsules.

5. Local-First, Your Data Stays Yours

• Your data, conversations, and credentials stay on your machines.
• No background syncing; encrypted manual sync and project transfer under your control.

---

Download & Install

Visit the Wisp Science download page and choose your OS and processor:

Platform	Package
Windows	Signed MSI / NSIS
macOS	Signed, notarized .dmg (Apple Silicon + Intel)
Linux	.deb / AppImage (x86_64 + aarch64)

After installing, open the bundled demo (no API key needed) to see a full RNA-seq trajectory. Then add a model in Settings → Models and start a project.

---

Comparison: RStudio vs VS Code vs Wisp Science

Feature	RStudio	VS Code + AI plugin	Wisp Science
Default R support	Yes	Requires extension	Built-in R kernel
Default Python support	Limited	Requires extension	Built-in Python kernel
AI assistance	None	Chat panel	Native agent: reads/writes files, runs code
Multi-language	R, Python	Multi-language	R, Python + MCP science databases
Persistent kernels	R only	None	Python and R persist across turns
Remote compute	None	None	SSH / WSL / GPU
Project memory	None	None	Full history, undoable
Database integration	None	None	~80 scientific databases

---

Next Steps

• 📥 Download Wisp Science
• 📖 Read the quick start: Quick Start
• ▶️ How to run FigureYa in Wisp Science: How to Run FigureYa in Wisp Science

Note: If you are still using VS Code, the original guide remains available as a reference, but we strongly recommend upgrading to Wisp Science for the full AI-powered research experience.