IRP

Name: IRP
Author: Noshadi-sec

作者：Noshadi-sec版本 0.6.0i★ –i在你的电脑上运行

注册名称io.github.Noshadi-sec/irp

安装

概览

作者提供的说明（英文原文）

Turn a research paper into working code and prove, claim by claim, how close it gets.

PopNx MCP 的三大支柱

这三项永远不会合并成一个分数：意见、测量和人气是不同的东西。

用户评价i暂无评价评价功能将在后续阶段与账户功能一起开放
我们的基础检查i1 项中通过 1 项依据公开信息检查；尚未扫描代码
使用情况i尚未统计★ –

它能帮你做什么

我们还没有为它写通俗的介绍。下面是我们从注册中心读取到的事实。

连接方式: 在你的电脑上运行
传输方式: stdio
分类: 搜索与 AI 记忆

安装

选择你使用的 AI 应用，然后复制设置。

安装之前

基础检查：1 项中通过 1 项
在你的电脑上运行，拥有你的用户账户的权限（文件、网络、应用）。
注册中心没有列出需要填写的密钥。

你需要

PyPI已安装 Python 的 uv（用于 uvx 命令）

你用的是哪个应用？

ClaudeClaude CodeCursorVS CodeChatGPT

在上方选择你的应用。我们会记住它，下次按钮会显示“安装到 …”。

打开 Claude Desktop → Settings → Developer → Edit Config，把下面的内容粘贴到 claude_desktop_config.json

{
  "mcpServers": {
    "irp": {
      "command": "uvx",
      "args": [
        "irp-agents",
        "mcp"
      ]
    }
  }
}

在终端中运行这条命令

claude mcp add irp -- uvx irp-agents mcp

在 Cursor 中打开

或者自己复制设置

粘贴到项目中的 .cursor/mcp.json，或 ~/.cursor/mcp.json

{
  "mcpServers": {
    "irp": {
      "command": "uvx",
      "args": [
        "irp-agents",
        "mcp"
      ]
    }
  }
}

浏览器会先询问是否打开应用，然后应用会显示设置并请你确认。密钥的值会留空，由你自己填写。

在 VS Code 中打开

或者自己复制设置

粘贴到项目中的 .vscode/mcp.json

{
  "servers": {
    "irp": {
      "type": "stdio",
      "command": "uvx",
      "args": [
        "irp-agents",
        "mcp"
      ]
    }
  }
}

浏览器会先询问是否打开应用，然后应用会显示设置并请你确认。密钥的值会留空，由你自己填写。

这个服务器在你的电脑上运行，而 ChatGPT 只能连接互联网上的服务器。请选择其他应用。

你的选择通过 Cookie 保存在这台设备上，无需账户。

这些设置是根据作者在注册中心的条目生成的。如果 README 的写法不同，请以 README 为准。

它能做什么

它在你的电脑上运行；安装后，工具列表会显示在你的 AI 应用中。

基础检查i

我们检查的是任何人都能看到的信息；我们没有阅读代码。不适用的检查项不计入总数。 · 检查于 2026年11月8日

✓ 通过
作者与仓库一致
注册名称已通过 GitHub 账户 Noshadi-sec 验证，该账户拥有此仓库
… 尚未检查
仍在维护
我们还没有从 GitHub 读取这个仓库
… 尚未检查
有许可证
我们还没有从 GitHub 读取这个仓库
– 不适用
端点有响应
它在你的电脑上运行，没有互联网端点
– 不适用
使用 OAuth 登录
在你的电脑上运行，改用本地设置
… 尚未检查
工具描述中没有隐藏指令
它在你的电脑上运行；我们只能读取互联网端点的工具列表

与“搜索与 AI 记忆”中的其他服务器比较i

星标数多于此分类中其他 5670 个服务器的 0%
发布到注册中心的时间比此分类中其他 5670 个服务器的 71% 更新
基础检查结果等于或优于此分类中其他 5670 个服务器的 100%

与注册中心里这个分类中所有活跃的服务器比较。只有当发布者拥有该仓库时，星标数才计入。

常见问题

IRP 安全吗？

它在 1 项适用的基础检查中通过了 1 项。我们没有阅读代码。在添加密钥或令牌之前，请先看看这个服务器要求哪些权限，并使用只具备所需权限的密钥。

它需要哪些密钥或代码？

注册中心没有列出必需的密钥，不过有些服务器会在首次使用时要求你登录。

它在我的电脑上运行还是在云端？

它在你的电脑上运行。你的 AI 应用会在需要时启动它，数据不会经过作者的服务器（除非它自己调用在线服务）。

它还在维护吗？

GitHub 仓库最后更新于–，最新版本（0.6.0）于17天前进入注册中心。

哪些应用可以使用它？

支持 MCP 的应用，例如 Claude Desktop、Claude Code、Cursor 和 VS Code。你的电脑上还需要有 Python (uv)。

有什么可以替代它？

同一分类中有：Context7, Codebase Memory, Exa

替代选择

同一分类中其他作者的服务器，按星标数排序。

如何看卡片：★ GitHub 星标i三大支柱：评价 · 检查 · 使用情况i更新 = 最新版本发布到注册中心的时间i

Context7

作者：upstash

Up-to-date docs for AI agents

电脑或云端★ 64,224MIT昨天更新

安装

Codebase Memory

作者：DeusData

Codebase knowledge graph for AI agents — 162 languages, sub-ms queries, 99% fewer tokens.

在你的电脑上运行★ 48,148MIT21天前更新

安装

Exa

作者：exa-labs

Connect AI agents to Exa for web search, content fetching, and multi-step research.

在云端运行★ 5,213MIT2个月前更新

安装

作者的 README

打开 README

按原文从作者的仓库复制。PopNx MCP 没有审查或背书其内容。HTML、图片和脚本已被移除。链接会离开本站。

English · 简体中文 · 日本語 · 한국어 · Español · Italiano

IRP: Build any paper. Prove every claim.

One MCP server and CLI that lets your coding agent turn a research paper into working code, run it, and prove claim by claim how close it gets.

Give it a paper. Get back working code and a report you can check without trusting the agent.

Quick start · How IRP works · The report · What you can reproduce · FAQ · Documentation

uvx irp-agents setup

[!IMPORTANT]

Get IRP from Noshadi-sec/irp or the

irp-agents package on PyPI. Follow the

installation and release links from these official sources. A similar

repository name or copied README does not establish that a download is official.

Found a paper with no code, or code you cannot trust? Ask your agent to build it with IRP. The agent reads the paper, records every number it claims, writes the implementation, runs it and compares the results. IRP does the fetching, running and record keeping, and refuses the shortcuts: a number typed by hand never counts, a shortened run never counts, and the report is written only from the evidence log.

IRP connects your agent (Claude Code, Codex or Cursor) to tools for paper intake, environments, runs, measurements and verdicts. You can use the same tools from your terminal. Everything runs locally; IRP never calls a model, and data goes only where you send it.

Quick start

Set up your agent

With uv installed, run:

uvx irp-agents setup

Choose your agents, review the proposed changes, and approve them. Setup adds IRP's MCP server and its workflow skill, with backups of existing configuration. Restart your agent afterward.

Setup supports Claude Code, Codex and Cursor. Other MCP clients can use manual registration.

Ask your agent

Build arXiv 2501.12345 and check Table 2.

You get back a working code folder plus a report: every claim, the paper's number, your number, and a verdict.

Use the terminal

Every tool is also a command. Fetch a paper into a new study and read its outline:

irp fetch 2501.12345 --study ~/studies/example

irp section index --study ~/studies/example

To install the irp command for regular use:

uv tool install irp-agents

irp --help

See the CLI guide for the full workflow from the terminal.

Check readiness

irp doctor

Doctor checks Python, uv, git, Docker, GPUs, disk space and your agents' registrations, and changes nothing.

Update IRP

IRP changes quickly. For an installed command:

irp update

With uvx, refresh your agent setup with uvx irp-agents@latest setup, review the plan, and restart your agent.

How IRP works

Your agent thinks; IRP acts and keeps the receipts. The agent reads the paper, writes code and decides what to run. Every action goes through one core, so the CLI and the MCP server behave the same, and every step lands in an append-only, hash-chained evidence log. The report is built from that log alone.

Stage	Tool	What it does
Intake	`fetch_paper`	Downloads the arXiv LaTeX source or the PDF, hashes it, splits it into sections
Intake	`get_section`, `get_page_image`	Returns one section's text with file and line numbers, or one page as an image for equations and figures
Claims	`add_claim`, `list_claims`	Records each claim: metric, value, where it appears, and the success bar
Plan	`add_guess`	Records a missing detail, the choice made, and why
Plan	`map_method`	Links an equation or algorithm step to a file and line
Build	`create_env`	Makes a uv environment or Docker image and writes the lock file
Run	`run`, `run_status`	Runs one command; saves output, exit code, time, memory, seed, commit and files
Run	`run_remote`	Same on a paid GPU provider, after a cost estimate and approval
Measure	`extract_metrics`	Pulls numbers from logs and result files, with file and line
Judge	`compare`, `add_note`	Puts claim and measurement side by side and gives one of four verdicts
Report	`report`, `verify`, `study_status`	Writes report.md and report.json from the log; rechecks the hash chain

The workflow skill gives the agent six hard rules, and the tools enforce them wherever they can:

Never type, estimate or copy a number as a result.
Never change the paper's method to make a run succeed; record a guess instead.
Report "could not run" with the blocking error instead of pushing through.
Run the smallest faithful version first; shortened runs never count.
Record every claim before writing any code.
Ask before any paid compute.

The report

The report is the product: a person must be able to check every verdict without trusting the agent. It has five parts: the claim table, four verdicts (reproduced, close, not reproduced, could not run), the guess list, the method map from equations to code lines, and provenance (arXiv ID and version, file hashes, git commit, lock files, GPU, seeds, every command with its exit code, and a verify command that rechecks the hash chain).

The agent cannot finish until every claim has a verdict with evidence.

Example report.md (made-up numbers):

# Reproduction report: Example Paper (arXiv 2501.12345v2)

Result: 2 reproduced, 1 close, 1 could not run (of 4 claims)

Claim	Where	Paper	Ours	Verdict
c1 Test accuracy	Table 2, row 3	94.1%	93.8% (3 seeds, std 0.2)	close
c2 Training loss	Figure 4	0.31	0.31 (3 seeds)	reproduced
c3 Speed-up	Section 5.2	2.0x	2.1x	reproduced
c4 ImageNet top-1	Table 3	81.2%	not run: needs 8 GPUs for 2 days	could not run

Guesses

g1 Warmup length not stated: used 5 epochs. Why: as in cited paper [12].

Method map

Eq. 3 (loss) -> code/src/loss.py:41
Algorithm 1, line 4 -> code/src/train.py:88

Provenance

Paper: arXiv 2501.12345v2, source and PDF hashes
Code: git commit 3f2a9c1 (clean)
Commands logged: 37, 37 with an exit code
Evidence log: 112 records, head sha256:…, verified
Verify: irp verify --study ~/studies/example

See evidence and report for every field.

What you can reproduce

IRP requires Python 3.11+ on macOS or Linux. Additional tools depend on the paper:

Need	What IRP uses	Requirements and guide
Paper intake	arXiv LaTeX source first, the PDF as fallback, page images for math and figures	Network access to arXiv; paper intake
Better PDF structure	Docling for text and tables	Optional `irp-agents[docling]`; paper intake
Environments	uv virtual environments with lock files, or Docker images for isolation	uv, or Docker; environments
Compute	Your CPU or GPU; Modal GPU sandboxes after a cost estimate and approval (experimental)	Optional `irp-agents[modal]`; compute
Agents	Claude Code, Codex, Cursor; any MCP client by hand	installation

IRP supports computational papers: machine learning, algorithms and simulations. It does not support wet-lab work or anything weapons-related.

Benchmark

Five light papers — official code exists, each runs on a CPU in minutes — rebuilt from the paper alone and checked claim by claim. Every number comes from a verified evidence log; re-check any study with irp verify --study benchmarks/studies/ .

Paper	Claims	Result
Fashion-MNIST (1708.07747)	3	1 reproduced, 1 close, 1 not reproduced
KMNIST (1812.01718)	2	both reproduced
MNIST-1D (2011.14439)	1	reproduced
fastText (1607.01759)	1	reproduced (92.86% vs 92.5%)
QMNIST (1905.10498)	1	reproduced (2.95% vs 2.95%)

Across 8 claims: 6 reproduced, 1 close, 1 not reproduced (87.5% reproduced or close), ~160s and $0 of paid compute per paper. Selection criteria and the full set are in benchmarks/golden/papers.json; the protocol and summarize.py are in benchmarks/.

If you build a paper with IRP, share the report in an issue with the paper, your question, and what you found.

FAQ

Which agents can use IRP?

Any agent that supports local MCP servers. Setup configures Claude Code, Codex and Cursor; other clients can use manual registration.

Does IRP call an AI model or upload my paper?

No. Your agent is the brain; IRP never calls a model and has no API key. It downloads papers from arXiv, installs packages through uv or Docker, and runs code on your machine. Your agent receives the tool results, and its model provider has its own data policy.

Is running a paper's code safe?

Running code is running code. A uv virtual environment is not a sandbox: the code runs with your permissions. Use create_env with backend: "docker" for code you did not write or do not trust, and runs mount only the study's code directory. See security.

What does "not reproduced" mean?

That this run, with the recorded guesses, did not reach the paper's number within the stated bar. The report says "not reproduced by this run" and lists every guess. It is evidence about one attempt, not a verdict on the paper; contact the authors before publishing it.

The paper has official code. Should I use IRP?

IRP builds missing code from the paper. To check that existing code reproduces its README's numbers, a tool such as paper-repro fits better. IRP is still useful for testing a paper by hiding its code and comparing your rebuild with it afterwards.

Can IRP spend money?

Only on paid GPU providers, only after you record a price and a budget at a terminal with irp budget, and only after you approve the estimate the agent shows you. The run's time limit equals the estimated hour

只显示了开头部分；其余内容请在仓库中阅读

IRP

概览