Paper Pilot

Your AI's research copilot.

Your AI Googles when you say "research." Paper Pilot actually searches real academic databases, downloads the PDFs, reads them cover to cover, and gives you evidence with citations.

CI PyPI License: MIT Python 3.11+ GitHub stars

---

Quick start

uvx paper-pilot

That's it. Or install it:

pip install paper-pilot

---

How it works

graph LR
    A[Prompt] --> B[Search 6 databases]
    B --> C[Resolve OA PDFs]
    C --> D[Download & read]
    D --> E[Extract evidence]
    E --> F[Render figures]
    F --> G[Markdown report]
    G --> H[Zotero sync]

One prompt. Six academic databases. Real PDFs. Real citations.

Research retrieval-augmented generation, deep-read the top papers, and compare the methods.

Your AI will:

1. Search Semantic Scholar, OpenAlex, arXiv, Crossref, and Europe PMC
2. Find the open-access PDFs, not just abstracts
3. Download and read them cover to cover
4. Extract evidence chunks with source attribution
5. Render specific pages so it can see the figures and tables
6. Write a structured Markdown report
7. Save everything into your Zotero library

---

vs. alternatives

	ChatGPT Deep Research	Gemini Deep Research	Perplexity Pro	Paper Pilot
Reads actual PDFs	Web summaries	Web summaries	Web summaries	Full text extraction
Figures and tables	Text only	Text only	Text only	Page rendering to PNG
Your library	Locked in their UI	Locked in Google	Locked in Perplexity	Syncs to Zotero
Sources	Generic web search	Generic web search	Web search	6 academic databases
Cost	$200/month	$20/month	$20/month	Free, MIT licensed
Your data	Their cloud	Their cloud	Their cloud	Your machine
Open source	No	No	No	Yes

---

MCP client setup

Claude Desktop

Add to your claude_desktop_config.json:

{
  "mcpServers": {
    "paper-pilot": {
      "command": "uv",
      "args": ["--directory", "/path/to/paper-pilot", "run", "paper-pilot"],
      "env": {
        "OPENALEX_EMAIL": "[email protected]",
        "UNPAYWALL_EMAIL": "[email protected]",
        "ZOTERO_LOCAL": "true",
        "SCIHUB_ENABLED": "false"
      }
    }
  }
}

Claude Code

claude mcp add --scope user paper-pilot -- uv --directory /path/to/paper-pilot run paper-pilot

Codex

Add to ~/.codex/config.toml:

[mcp_servers.paper_pilot]
command = "uv"
args = ["--directory", "/path/to/paper-pilot", "run", "paper-pilot"]

[mcp_servers.paper_pilot.env]
OPENALEX_EMAIL = "[email protected]"
ZOTERO_LOCAL = "true"

Streamable HTTP mode

paper-pilot --transport streamable-http --host 127.0.0.1 --port 8000

---

Tools

Tool	What it does
research_topic	Full pipeline: search, download, report, Zotero sync
deep_read_topic	Everything above + full-text extraction with evidence chunks
render_pdf_pages	PDF pages to PNG for figure and table inspection
search_literature	Fine-grained multi-source academic search
find_similar_papers	Related work expansion from a seed paper
inspect_open_access_pdf	OA availability check and PDF preview
extract_local_pdf_text	Text extraction from any local PDF
search_scihub	Search Sci-Hub by DOI, title, or keyword (opt-in)
download_scihub_paper	Download a paper via Sci-Hub by DOI (opt-in)
search_libgen	Supplementary shadow library search
healthcheck	Verify all connections are up

---

Sci-Hub integration (opt-in)

Sci-Hub access is disabled by default. To opt in:

SCIHUB_ENABLED=true

Once enabled, use search_scihub and download_scihub_paper directly, or pass include_scihub=True to research_topic / deep_read_topic for automatic fallback.

Disclaimer: Sci-Hub integration is provided strictly for educational and research purposes. Users are solely responsible for compliance with applicable laws and institutional policies.

---

Who uses this

PhD students that don't want to spend a week on a literature review. Point it at your thesis topic, get back a structured comparison with real citations and the PDFs already in Zotero.

Research labs that want to scan preprints weekly and auto-file them. Run research_topic on a schedule and keep your group library current.

AI builders that need their agents to work with real academic papers instead of web scraping snippets.

---

Configuration

[email protected]        # Required for polite API access
[email protected]       # Required for OA resolution
SEMANTIC_SCHOLAR_API_KEY=             # Optional, higher rate limits

# Local Zotero
ZOTERO_LOCAL=true
ZOTERO_LIBRARY_TYPE=user

# Web Zotero API (alternative)
ZOTERO_LIBRARY_ID=
ZOTERO_API_KEY=

# Sci-Hub (disabled by default)
SCIHUB_ENABLED=false

# Storage
PAPER_PILOT_DATA_DIR=./data

# Institutional networks
HTTP_PROXY=
HTTPS_PROXY=
SSL_CERT_FILE=

---

Project structure

src/paper_pilot/
  server.py              MCP tools and pipeline orchestration
  config.py              Environment and settings
  services/
    academic.py          Multi-source scholarly search
    open_access.py       OA resolution and PDF downloads
    scihub.py            Sci-Hub paper resolution (opt-in)
    deep_read.py         Full-text extraction and page rendering
    zotero.py            Local and web Zotero integration
    reporting.py         Markdown report generation
    libgen.py            Supplementary LibGen support

Architecture details: docs/ARCHITECTURE.md

---

For AI agents

• AGENTS.md -- shared operating guide
• CLAUDE.md -- Claude Desktop and Claude Code setup
• CODEX.md -- Codex setup
• docs/CLIENTS.md -- side-by-side client comparison

---

Contributing

PRs welcome. The most impactful areas:

• New scholarly source adapters
• Better OA resolution logic
• PDF parsing improvements
• More MCP client configs

See CONTRIBUTING.md.

---

Disclaimer

This tool is designed for academic research and educational purposes only. Open-access features use only legal, publicly available sources. Sci-Hub and LibGen integrations are disabled by default and provided as opt-in features.

---

License

MIT. Do whatever you want with it.

If this helps your research, star the repo and tell a colleague.