Context7
Up-to-date docs for AI agents
Beschreibung vom Entwickler (englisches Original)
Turn a research paper into working code and prove, claim by claim, how close it gets.
Diese drei werden nie zu einer Punktzahl zusammengefasst: Meinungen, Messungen und Beliebtheit sind verschiedene Dinge.
Für diesen Server haben wir noch keine Zusammenfassung in einfacher Sprache geschrieben. Unten stehen die Fakten, die wir aus der Registry gelesen haben.
Wähle die KI-App, die du nutzt, und kopiere die Einstellungen.
Du brauchst
Welche App nutzt du?
Wähle oben deine App. Wir merken sie uns, damit der Button beim nächsten Mal „In … installieren“ anzeigt.
Öffne Claude Desktop → Settings → Developer → Edit Config und füge dies in claude_desktop_config.json ein
{
"mcpServers": {
"irp": {
"command": "uvx",
"args": [
"irp-agents",
"mcp"
]
}
}
}Führe das in deinem Terminal aus
claude mcp add irp -- uvx irp-agents mcpIn .cursor/mcp.json in deinem Projekt oder in ~/.cursor/mcp.json einfügen
{
"mcpServers": {
"irp": {
"command": "uvx",
"args": [
"irp-agents",
"mcp"
]
}
}
}Dein Browser fragt, ob er die App öffnen darf, dann zeigt die App die Einstellungen und bittet dich um Bestätigung. Schlüsselwerte bleiben leer, damit du sie selbst einträgst.
In .vscode/mcp.json in deinem Projekt einfügen
{
"servers": {
"irp": {
"type": "stdio",
"command": "uvx",
"args": [
"irp-agents",
"mcp"
]
}
}
}Dein Browser fragt, ob er die App öffnen darf, dann zeigt die App die Einstellungen und bittet dich um Bestätigung. Schlüsselwerte bleiben leer, damit du sie selbst einträgst.
Dieser Server läuft auf deinem Computer, und ChatGPT kann sich nur mit Servern im Internet verbinden. Wähle eine andere App.
Deine Auswahl bleibt in einem Cookie auf diesem Gerät; kein Konto nötig.
Diese Einstellungen werden aus dem Registry-Eintrag des Entwicklers erzeugt. Wenn in der README etwas anderes steht, folge der README.
Läuft auf deinem Computer; die Tool-Liste erscheint nach der Installation in deiner KI-App.
Wir prüfen Informationen, die jeder sehen kann; den Code haben wir nicht gelesen. Checks, die nicht zutreffen, werden nicht mitgezählt. · Geprüft am 7. Nov. 2026
Der Registry-Name wurde mit dem GitHub-Konto Noshadi-sec bestätigt, dem das Repo gehört
Dieses Repo haben wir noch nicht von GitHub gelesen
Dieses Repo haben wir noch nicht von GitHub gelesen
Dieser Server läuft auf deinem Computer; es gibt keinen Endpunkt im Internet
Läuft auf deinem Computer und nutzt stattdessen lokale Einstellungen
Läuft auf deinem Computer; Tool-Listen können wir nur von Endpunkten im Internet lesen
Verglichen mit allen aktiven Servern dieser Kategorie in der Registry. Sterne zählen nur, wenn dem Herausgeber das Repo gehört.
Er hat 1 von 1 zutreffenden Basischecks bestanden. Den Code haben wir nicht gelesen. Bevor du einen Schlüssel oder Token hinzufügst, sieh nach, welche Zugriffe der Server verlangt, und nutze einen Schlüssel, der nur die Berechtigungen hat, die du brauchst.
Die Registry nennt keinen erforderlichen Schlüssel, manche Server bitten dich aber bei der ersten Nutzung, dich anzumelden.
Er läuft auf deinem Computer. Deine KI-App startet ihn bei Bedarf, und die Daten laufen nicht über die Server des Entwicklers (außer der Server ruft selbst einen Onlinedienst auf).
Das GitHub-Repo wurde zuletzt – geändert, und die neueste Version (0.6.0) kam vor 16 Tagen in die Registry.
Apps, die MCP unterstützen, etwa Claude Desktop, Claude Code, Cursor und VS Code. Außerdem brauchst du Python (uv) auf deinem Computer.
In derselben Kategorie: Context7, Codebase Memory, Exa
Server derselben Kategorie von anderen Entwicklern, sortiert nach Sternen.
So liest du eine Karte:★ GitHub-SterneiDrei Säulen: Bewertungen · Checks · NutzungiAktualisiert = neueste Version in der Registry veröffentlichti
Up-to-date docs for AI agents
Codebase knowledge graph for AI agents — 162 languages, sub-ms queries, 99% fewer tokens.
Connect AI agents to Exa for web search, content fetching, and multi-step research.
Unverändert aus dem Repo des Entwicklers übernommen. PopNx MCP hat den Inhalt weder geprüft noch empfiehlt es ihn. HTML, Bilder und Skripte wurden entfernt. Links führen von dieser Website weg.
English · 简体中文 · 日本語 · 한국어 · Español · Italiano
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.
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.
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.
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.
irp doctor
Doctor checks Python, uv, git, Docker, GPUs, disk space and your agents' registrations, and changes nothing.
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.
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:
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 |
irp verify --study ~/studies/example
See evidence and report for every field.
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.
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.
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
Nur der Anfang wird gezeigt; den Rest findest du im Repo