Wisp ScienceScientific workspace
中文
OPEN SOURCE · LOCAL-FIRST · SCIENTIFIC AGENT WORKSPACEVOL. 01 / v0.14.0

Wisp ScienceFrom research question to reproducible artifact.

Wisp Science brings papers, local data, Python, biological databases, and agent workflows into one traceable workspace—on desktop or through a headless CLI.

Download the latest releaseExplore the research workflow

Bring your own model. Keep project state local.

License
Apache-2.0
Interfaces
Desktop / CLI
Status
Active preview
RESEARCH RECORD · WS-014Analysis trace preserved
QUESTION

Which pathways distinguish the treatment-responsive cell population?

responsive population
01Evidence

12 papers · 3 local datasets

02Method

Scanpy workflow · persistent Python

03Result

Marker table · UMAP · methods note

Fig. 01 — Evidence, code, tool calls, and exported artifacts remain linked to the same session.
0129

bundled research workflows

02~80

biological database clients

03Python

persistent local kernel

04Local

sessions and artifacts

A TRACEABLE RESEARCH LOOP

Move from discovery to communication without losing the method.

Each stage remains connected to its evidence, code, tools, and generated outputs, so the final report is a record of the work—not just an answer.

  1. 01

    Discover

    Search papers and biological databases, inspect local files, and assemble evidence around the question.

  2. 02

    Analyze

    Run Python, shell, MCP tools, and domain SKILL workflows while preserving computational state.

  3. 03

    Verify

    Review plans before execution, inspect intermediate files, and trace every artifact back to inputs and code.

  4. 04

    Communicate

    Export figures, tables, methods, citations, and narrative reports from one coherent research record.

REPRODUCIBILITY, BUILT INTO THE SESSION

Evidence, code, and context—kept in the same research record.

Wisp Science treats outputs as artifacts with provenance. Figures and tables can retain their producing code, logs, inputs, and environment details.

01

@ file references

Attach the file already open or any file in the current project without breaking the research context.

02

Plan approval

Let the agent propose a multi-step method, pause for review, and only continue after approval.

03

Artifact provenance

Open figures and CSVs with their code, logs, input files, and download actions alongside them.

PROVENANCE LEDGERWS / TRACE-014
01Inputcounts_matrix.h5adsha256 verified
02Methodscanpy_marker_workflow29 steps logged
03Figureumap_response.pngcode attached
04Reportmethods_and_results.mdcitations linked
REPRODUCIBLE

SCIENTIFIC COMPUTING, NOT GENERIC CHAT

The local tools a research agent actually needs.

PY01

Persistent Python REPL

A uv-managed kernel keeps state across turns for iterative analysis, plotting, and debugging.

MCP02

Biological databases

Built-in bio MCP servers expose around 80 database clients as tools the agent can call.

SK03

29 research SKILLs

Workflows cover literature, proteins, single-cell analysis, cheminformatics, figures, and remote compute.

FS04

Files and shell

Read, write, search, edit, grep, and shell tools operate inside the boundary of the local project.

DB05

Local project history

Projects, messages, session frames, settings, and artifacts are stored locally in SQLite.

CTX06

Long-session context

Layered compaction keeps extended investigations moving without discarding the research trail.

RESEARCH PLATES

One workspace, different scientific questions.

Compose database retrieval, local computation, domain workflows, and structured writing around the task at hand.

PLATE 0101 / 04

Single-cell analysis

Run Scanpy or scVI-style workflows, annotate populations, and export UMAPs, marker tables, and methods notes.

UMAP · MARKERS · METHODS
PLATE 0202 / 04

Protein structure

Fetch sequences and structures, combine AlphaFold-, Boltz-, or OpenFold-style skills, and draft structured interpretation.

SEQUENCE · STRUCTURE · REPORT
PLATE 0303 / 04

Cheminformatics

Search ChEMBL or PubChem, compare activity data, calculate properties, and prepare SAR-style tables.

ACTIVITY · PROPERTIES · SAR
PLATE 0404 / 04

Literature and writing

Search PubMed or Semantic Scholar, inspect PDFs, draft discussion sections, and keep citations with the text.

PAPERS · CITATIONS · DRAFT

DEMO ATLAS

Inspect a complete agent trace before using API credits.

Bundled read-only sessions expose the path from question to evidence and exported output for common life-science tasks.

01CRISPR Screen

Raw counts → hit calling → visualization report

02Enzyme Engineering

Sequence analysis → mutation proposals → activity reasoning

03Extremophile

Metagenomic search → cross-database functional annotation

04Immunotherapy

Tumor immunology literature → evidence map → hypotheses

YOUR MODEL. YOUR COMPUTE. YOUR PROJECT.

Local-first by design, explicit about the network boundary.

Raw files, sessions, settings, and artifacts stay with the local project. Prompts and model responses still pass through the provider you configure, and database calls follow each remote service’s policy.

Model providers
OpenAI-compatible APIs, OpenAI Responses, and Anthropic
Execution
Desktop application, headless CLI, SSH, and Modal workflows
Local state
SQLite project history and operating-system keychain storage

LATEST RELEASE

Install Wisp Science and begin with a local project.

Release files are served by GitHub. Package checksums are published here so each download can be verified before use.

Current releasev0.14.0

Published Jul 16, 2026. macOS packages are signed and notarized; Windows packages remain unsigned and may trigger SmartScreen.

Read release notes

QUESTIONS BEFORE YOU BEGIN

What Wisp Science is—and where its boundaries are.

01Is Wisp Science a new model?

No. It is an open-source desktop and CLI application that uses the model provider and credentials you configure.

02How is it different from a generic AI assistant?

It can execute local Python, shell and file tools, call MCP databases, follow domain SKILL workflows, and retain artifact provenance.

03Does research data stay private?

Project files, sessions, artifacts, and settings remain local. Prompts and responses still pass through your configured model provider.

04Is it production-stable?

Wisp Science is an active preview for local scientific workflows. Review critical methods and outputs, and check release notes for current signing and update status.