Metadata-Version: 2.4
Name: scibudy
Version: 0.2.3
Summary: Codex-native scientific research expansion assistant with scholarly search, library management, and local semantic analysis
Author-email: ONEMULE <xinluo116@gmail.com>
License-Expression: Apache-2.0
Project-URL: Homepage, https://github.com/ONEMULE/scibudy
Project-URL: Repository, https://github.com/ONEMULE/scibudy
Project-URL: Issues, https://github.com/ONEMULE/scibudy/issues
Project-URL: Documentation, https://github.com/ONEMULE/scibudy/tree/main/docs
Project-URL: Changelog, https://github.com/ONEMULE/scibudy/blob/main/CHANGELOG.md
Keywords: codex,mcp,research,literature-search,scientific-workflow,semantic-retrieval
Classifier: Development Status :: 4 - Beta
Classifier: Intended Audience :: Science/Research
Classifier: Intended Audience :: Developers
Classifier: Operating System :: MacOS
Classifier: Operating System :: POSIX :: Linux
Classifier: Programming Language :: Python :: 3
Classifier: Programming Language :: Python :: 3.10
Classifier: Programming Language :: Python :: 3.11
Classifier: Programming Language :: Python :: 3.12
Classifier: Topic :: Scientific/Engineering
Classifier: Topic :: Software Development :: Libraries
Classifier: Topic :: Utilities
Requires-Python: >=3.10
Description-Content-Type: text/markdown
License-File: LICENSE
Requires-Dist: mcp==1.27.0
Requires-Dist: httpx==0.28.1
Requires-Dist: pydantic<3.0.0,>=2.13.0
Requires-Dist: pydantic-settings<3.0.0,>=2.13.1
Requires-Dist: python-dotenv<2.0.0,>=1.2.2
Requires-Dist: beautifulsoup4<5.0.0,>=4.12.3
Requires-Dist: openai<3.0.0,>=1.102.0
Requires-Dist: pypdf<7.0.0,>=5.9.0
Provides-Extra: dev
Requires-Dist: pytest==9.0.3; extra == "dev"
Requires-Dist: build>=1.2.2; extra == "dev"
Provides-Extra: docs
Requires-Dist: mkdocs>=1.6.1; extra == "docs"
Requires-Dist: mkdocs-material>=9.6.14; extra == "docs"
Provides-Extra: release
Requires-Dist: build>=1.2.2; extra == "release"
Requires-Dist: twine>=6.1.0; extra == "release"
Dynamic: license-file

# Scibudy

[![CI](https://github.com/ONEMULE/scibudy/actions/workflows/ci.yml/badge.svg)](https://github.com/ONEMULE/scibudy/actions/workflows/ci.yml)
[![Docs](https://github.com/ONEMULE/scibudy/actions/workflows/docs.yml/badge.svg)](https://github.com/ONEMULE/scibudy/actions/workflows/docs.yml)
[![Release Check](https://github.com/ONEMULE/scibudy/actions/workflows/release-check.yml/badge.svg)](https://github.com/ONEMULE/scibudy/actions/workflows/release-check.yml)

Scibudy is a Codex-native scientific research expansion assistant for scholarly search, library management, full-text ingestion, and local semantic analysis.

Scibudy combines:

- a local MCP server for Codex
- a shell-first CLI
- a browser management UI
- a layered install system for CPU-first and GPU-extended deployments

中文简介：

Scibudy 是一个面向 Codex 的科研增强助手，提供学术检索、文献库管理、全文分析和本地高质量语义检索能力。它既可以作为 MCP 工具，也可以作为独立 CLI 和本地管理界面使用。

## Status

- License: Apache-2.0
- Release posture: stable `v0.x`
- Primary platforms: Linux and macOS
- Full local GPU path: Linux + NVIDIA first

## Quick links

- English docs: [docs/en/index.md](docs/en/index.md)
- 中文文档: [docs/zh/index.md](docs/zh/index.md)
- Prerequisites: [docs/en/prerequisites.md](docs/en/prerequisites.md) / [docs/zh/prerequisites.md](docs/zh/prerequisites.md)
- Codex MCP setup: [docs/en/codex-setup.md](docs/en/codex-setup.md) / [docs/zh/codex-setup.md](docs/zh/codex-setup.md)
- Usage examples: [docs/en/examples.md](docs/en/examples.md) / [docs/zh/examples.md](docs/zh/examples.md)
- Architecture: [docs/en/architecture.md](docs/en/architecture.md) / [docs/zh/architecture.md](docs/zh/architecture.md)
- Support matrix: [docs/en/support-matrix.md](docs/en/support-matrix.md) / [docs/zh/support-matrix.md](docs/zh/support-matrix.md)
- Contributing: [CONTRIBUTING.md](CONTRIBUTING.md)
- Security: [SECURITY.md](SECURITY.md)
- Support: [SUPPORT.md](SUPPORT.md)
- Roadmap: [ROADMAP.md](ROADMAP.md)
- Changelog: [CHANGELOG.md](CHANGELOG.md)

## Installation

### Before you install

For most new users, the real prerequisites are only:

- Node.js 18+
- Python 3.10+

Read more:

- English: [docs/en/prerequisites.md](docs/en/prerequisites.md)
- 中文: [docs/zh/prerequisites.md](docs/zh/prerequisites.md)

### Unified installer

```bash
npx scibudy-install --profile base
```

Profiles:

- `base`: search, library management, UI, Codex config
- `analysis`: base + analysis-oriented runtime conventions
- `gpu-local`: local GPU model environment and warm flow
- `full`: full bootstrap for a Linux GPU workstation

### Source install

```bash
git clone git@github.com:ONEMULE/scibudy.git
cd scibudy
python3 -m venv .venv
. .venv/bin/activate
python -m pip install -e .[dev]
scibudy bootstrap --profile base --install-codex
```

## Runtime commands

Primary command aliases:

- `scibudy`
- `scibudy-mcp`
- compatibility aliases: `research-cli`, `research-mcp`

Examples:

```bash
scibudy search "simulation-based calibration" --mode general
scibudy collect "simulation-based calibration" --target-dir ~/Desktop/sbc-library
scibudy analysis-settings
scibudy ingest-library <library_id>
scibudy search-evidence <library_id> calibration
scibudy profiles
scibudy workflow "calibration methods in simulation-based inference" --limit 50 --topic "calibration in simulation-based inference"
scibudy workflow "calibration methods in simulation-based inference" --dry-run
scibudy workflow "calibration methods in simulation-based inference" --quality-mode fast
scibudy security-audit
scibudy doctor --install-readiness
scibudy synthesize-library <library_id> "causal inference robustness" --profile general
scibudy synthesize-library <library_id> "calibration in simulation-based inference" --profile sbi_calibration
scibudy ui --open
```

For Codex and other agents, prefer the high-level workflow entrypoint when you want the whole research loop:

```text
Use research/research_workflow with query="calibration methods in simulation-based inference", mode="general", limit=50, synthesize=true.
```

Use lower-level tools such as `search_literature`, `collect_library`, `ingest_library`, and `build_research_synthesis` when you need manual control over each step.

Use `dry_run=true` when an agent should preview writes and planned steps before executing. Use `quality_mode=fast` for low-cost exploration, `standard` for the normal workflow, and `deep` when missing full text or unsupported claims require stricter follow-up.

For safer agent automation, run `scibudy security-audit` and `scibudy doctor --install-readiness` before delegating long-running research workflows.

## Domain profiles

Domain profiles do **not** limit Scibudy's search scope or providers. Search remains general and multi-source by default.

Profiles only tune full-text synthesis: section weighting, evidence markers, unsupported-claim detection, and risk flags.

- `general`: default all-domain synthesis profile.
- `auto`: chooses a synthesis profile from the topic while preserving general search.
- `sbi_calibration`: an example preset for simulation-based inference calibration workflows.

For more examples and Codex prompt patterns:

- English: [docs/en/examples.md](docs/en/examples.md)
- 中文: [docs/zh/examples.md](docs/zh/examples.md)

## Local model stack

The highest-quality local retrieval path currently uses:

- `Qwen/Qwen3-Embedding-4B`
- `Qwen/Qwen3-Reranker-4B`

Recommended workflow:

```bash
scibudy install-local-models
scibudy warm-local-models --background
```

See:

- English: [docs/en/gpu-local.md](docs/en/gpu-local.md)
- 中文: [docs/zh/gpu-local.md](docs/zh/gpu-local.md)

## Repository layout

```text
research_mcp/   Python runtime, MCP server, CLI, analysis engine
web/            UI source and built assets
bin/            npm/bootstrap entrypoints
docs/           Bilingual project documentation
examples/       Copyable usage examples
scripts/        Release and smoke-check helpers
.github/        CI, templates, automation
```

## Open-source project standards

This repository is intentionally organized like a professional open-source library:

- documented install profiles
- release manifest and bootstrap state
- contributor and support policies
- issue/PR templates
- CI and packaging checks
- bilingual documentation for core user workflows

## Development

Core local checks:

```bash
make test
make build-ui
make package-check
make release-check
```

For deeper guidance:

- English: [docs/en/development.md](docs/en/development.md)
- 中文: [docs/zh/development.md](docs/zh/development.md)
