Stanford 团队发布 Paper2Agent,将研究论文转化为可复现结果的 MCP 服务器

MarkTechPost(RSS)·2026-09-17 05:59·56分钟前·Asif Razzaq
AI 导读

Stanford 团队(Jiacheng Miao、James Zou)的 Paper2Agent 于 2026 年 9 月 16 日发表在 Nature,可将论文及其代码库自动转换为 MCP 服务器,供 Claude Code 等 MCP 兼容智能体用自然语言调用论文方法。

MarkTechPost(RSS)
67AI 编辑部评分,满分 100

Stanford 团队发布 Paper2Agent,将研究论文转化为可复现结果的 MCP 服务器

2026-09-17 05:59· 56分钟前· Asif Razzaq
AI 导读

Stanford 团队(Jiacheng Miao、James Zou)的 Paper2Agent 于 2026 年 9 月 16 日发表在 Nature,可将论文及其代码库自动转换为 MCP 服务器,供 Claude Code 等 MCP 兼容智能体用自然语言调用论文方法。

Computational papers ship code that readers must clone, install, configure and debug. That cost keeps useful methods locked inside PDFs. A Stanford team led by Jiacheng Miao and James Zou proposes a fix. Paper2Agent was published in Nature on 16 September 2026. It converts a paper and its codebase into a Model Context Protocol (MCP) server. Any MCP-compatible agent, such as Claude Code, can then run the paper’s methods through natural language. The authors describe the result as a virtual corresponding author.

Is it deployable? Yes. The code is MIT-licensed and installs as a skill for Claude Code or Codex. Prebuilt AlphaGenome, Scanpy and TISSUE servers run on Hugging Face Spaces. A hosted version is also available at paper2agent.ai.

How the Pipeline Works

Paper2Agent runs on Claude Code’s agent SDK. A central orchestrator dispatches specialized sub-agents through 6 steps:

  1. Locate and download the codebase.
  2. An environment manager builds an isolated virtual environment.
  3. A tutorial scanner indexes usable tutorials.
  4. A tutorial executor runs them end to end and records reference outputs.
  5. A tool extractor turns tutorials into parameterized MCP tools, and a test verifier validates them.
  6. The orchestrator assembles validated tools into 1 MCP server.

The validation gate is strict. A tool passes only when expected files appear and numbers match within 3%. Figures must also match references by perceptual hash, with Hamming distance under 20. The verifier gets up to 6 attempts per function. Tools that keep failing are excluded from the final server.

Each server exposes 3 components. MCP tools wrap the paper’s methods as executable functions: MCP resources hold the manuscript, code links, datasets and figures. MCP prompts encode multi-step workflows, such as the correct Scanpy preprocessing order. The research team used Claude Sonnet 4 for all Paper2Agent applications.

Interactive Explainer

AlphaGenome Agent Results

For AlphaGenome, Paper2Agent built 22 tools in about 45 minutes for US $14. All 22 passed validation without human intervention. The team compared the agent with Claude Code plus repository access (Claude + Repo) and Biomni.

BenchmarkPaper2AgentClaude + RepoBiomni
15 tutorial-derived queries98.7 ± 1.3%82.7 ± 3.4%37.3 ± 4.0%
15 novel queries100.0 ± 0.0%78.7 ± 4.4%56.0 ± 3.4%
30 open-ended queries82.7 ± 2.4%56.7 ± 2.3%72.2 ± 2.2%

Results span 5 runs, graded by 2 human experts with 96.7% inter-rater agreement. On tutorial queries, median runtime fell 1.9× versus Claude + Repo and 3.1× versus Biomni. The gains persisted when the baseline was upgraded to Claude Opus 4.6.

The agent also re-examined an LDL cholesterol variant, chr1:109274968:G>T. It ranked SORT1 as the likely causal gene. The original AlphaGenome paper emphasized CELSR2 and PSRC1. GTEx shows significant liver eQTLs for all 3 genes. The research team say this shows how hard causal gene assignment is at such loci.

Scanpy, TISSUE and Scale Tests

The Scanpy agent received 7 validated tools in about 45 minutes for US $13. On 4 public datasets, it matched human researchers on cell counts, gene counts and top marker genes. A TISSUE agent reproduced human results on spatial transcriptomics data.

Scale tests covered 3 corpora with no manual cleanup:

  • 100 bioRxiv computational biology papers: 74 were agentified, and 593 of 599 proposed tools passed validation.
  • 300 questions: Paper2Agent scored 91.2%, versus 80.3% (Sonnet 4) and 86.3% (Sonnet 4.6) for Claude + Repo.
  • Cost per query: US $0.20 and 1.6 minutes, compared with US $0.38 and 4.3 minutes.
  • 10 non-biology papers, including TabPFN, SAM 2 and SAELens: 98.1% accuracy on 42 execution tasks.
  • 26 data-focused papers: resource layer 89.0% versus 82.0% for browser use, 34× cheaper and 15× faster.

Paper2Agent also rejected 100% of out-of-scope queries in a permuted benchmark. It recovered from injected dependency, file-path, typo and deprecated API failures.

Paper Agents Collaborating

The research team connected 3 agents: AlphaGenome, an MPRA-coupled scCRISPRi screen and a CD4+ T cell Perturb-seq dataset. AlphaGenome flagged GPR137 at psoriasis locus rs887314, with an RNA-seq quantile score of 0.997. The AI co-scientist proposed 10 validation strategies, and a researcher picked signature correlation.

Only GPR137 knockdown matched the CRE perturbation signature. The match appeared under stimulation: Spearman 0.613 at Stim8hr and 0.630 at Stim48hr. BAD and 3 other candidates showed no significant correlation. A second study paired AlphaGenome with an ADHD GWAS and nominated rs1626703 among 209 candidates. That hypothesis still needs experimental validation.

Key Takeaways

  • Paper2Agent converts papers and repos into tested MCP servers with tools, resources and prompts.
  • The AlphaGenome agent took about 45 minutes, cost US $14, and scored 100% on novel queries.
  • 74 of 100 bioRxiv papers were agentified, with 593 of 599 tools validated.
  • 3 paper agents jointly supported GPR137 as the probable psoriasis causal gene.
  • The code is MIT-licensed, with prebuilt MCP servers on Hugging Face Spaces.

来源:MarkTechPost(RSS)· marktechpost.com