frontpage.
newsnewestaskshowjobs

Made with ♥ by @iamnishanth

Open Source @Github

fp.

Open in hackernews

MCP servers mass-forked and republished – supply-chain attack vector

1•ultrafox42•1h ago
I'm the developer of an MCP server (AiDex — code indexing with Tree-sitter, github.com/CSCSoftware/AiDex). I recently discovered that an organization called iflow-mcp is systematically forking hundreds of MCP servers on GitHub, republishing them under their own npm scope (@iflow-mcp/) and on PyPI, and distributing them through their own "marketplace" — without ever contacting the original authors.

But the real problem isn't the fork itself. The real problem is security.

---

*What's happening*

The forks are published under the pattern @iflow-mcp/originalauthor-projectname. The familiar name of the original creates trust — but the code the user actually gets is entirely under a third party's control. Nothing prevents that third party from modifying the code before publishing.

*Why MCP servers are particularly dangerous*

An MCP server is not a harmless plugin. By design, it has deep access:

- It reads your source code, navigates your filesystem, receives file contents - It can see .env files, API keys, SSH keys, credentials, and proprietary code - It communicates via stdio directly with the AI client — meaning it can manipulate tool responses - It often runs with the same permissions as the user

A trojanized MCP server could silently exfiltrate sensitive data while doing exactly what it's supposed to do — indexing or analyzing your code. You see the tool working normally. You don't see the data leaving.

Even worse: a manipulated server could feed the AI assistant deliberately false information — suggesting insecure code patterns, hiding vulnerabilities, or redirecting the AI to the wrong files.

*The broken chain of trust*

The problem is systemic:

1. Developer A publishes an MCP server as open source 2. Organization X forks it, republishes it under their own scope 3. A user finds the package in X's marketplace, recognizes the familiar name, installs it 4. The user believes they're running the original — but they're running code that X could have modified in any way

This is a classic supply-chain attack. And it works particularly well because MCP servers are new, and many users haven't yet learned to verify where their tools actually come from.

---

*What you can do as an MCP server developer*

1. *Origin check at startup:* Check your own package name (package.json, process.env.npm_package_name) and installation path (__dirname). If a foreign scope appears, display a clear warning:

    This appears to be an unofficial redistribution of [project].
   Official package: npm install [original-package]
   Repository: [original-url]
2. *Build signature:* Embed a hash or signed identifier during your build process. A republisher who doesn't replicate your exact build pipeline cannot reproduce it.

3. *Registry check:* A quick HTTP request at startup to the npm registry can reveal whether the package is running under a foreign scope.

*What you can do as a user*

- Always install MCP servers directly from the original project, never through third-party marketplaces - Check the npm scope: @iflow-mcp/cscsoftware-aidex is NOT the same as aidex-mcp - Look at GitHub to verify whether a repository is a fork and who the actual author is - Be especially cautious with MCP servers that have deep filesystem access

---

*Please share this post.* The more MCP developers and users know about this attack vector, the harder it becomes to exploit. The MCP ecosystem is growing rapidly — but without awareness of supply-chain security, we're building on sand.

If you're affected: speak up. Check whether your project appears under github.com/iflow-mcp/. The more voices, the sooner npm and GitHub will act.

Original project: https://github.com/CSCSoftware/AiDex Their fork on npm: https://www.npmjs.com/package/@iflow-mcp/cscsoftware-aidex

— Uwe Chalas, author of AiDex (aidex-mcp on npm)

Fact Check: Alec and Kaleb Are Alive and Well

https://www.shrinerschildrens.org/en/news-and-media/news/2026/03/fact-check-alec-and-kaleb-are-al...
1•rolph•30s ago•0 comments

Intel briefed that new supreme leader is probably gay

https://nypost.com/2026/03/16/us-news/trump-briefed-that-irans-new-supreme-leader-mojtaba-khamene...
1•nothrowaways•31s ago•0 comments

Browser extension that makes LLMs appear to run slowly (ChatGPT and Claude)

https://slowllm.lav.io/
2•teetaa•3m ago•0 comments

Show HN: What if AI agents can trade with each other

https://openstall.ai
1•ljhnick•4m ago•0 comments

Gitmore – Real-time engineering visibility from Git activity

https://gitmore.io
1•amitousablitou•7m ago•1 comments

I Built a Spy Satellite Simulator in a Browser. Here's What I Learned

https://www.spatialintelligence.ai/p/i-built-a-spy-satellite-simulator
2•cyrc•7m ago•0 comments

Ask HN: How do you manage your relationships?

1•anqer•8m ago•0 comments

The Situation Room by Polymarket Is Opening This Friday in DC

https://www.popville.com/2026/03/the-situation-room-polymarket-bar-washington-dc/
1•ryan_j_naughton•10m ago•0 comments

Ask HN: Can we please stop with the posts about Claude outages?

2•romanhn•11m ago•2 comments

A Mermaid Planning Tool for AI

https://relistan.com/mermaid-tool-for-ai
1•relistan•12m ago•0 comments

Towards a Physics Foundation Model

https://flowsnr.github.io/blog/physics-foundation-model/
1•e-topy•12m ago•0 comments

Procedural Planets

https://franpiaggio.github.io/planets/
1•memalign•13m ago•0 comments

The GPT Sexbot

https://tapestry.news/tech/chatgpt-adult-mode/
2•sygona•13m ago•0 comments

DOGE canceled NC Museum grant for HVAC systems after ChatGPT flagged it as DEI

https://myfox8.com/news/north-carolina/high-point/doge-canceled-high-point-museum-grant-for-hvac-...
7•cldwalker•14m ago•1 comments

Writing for Developers

https://www.manning.com/books/writing-for-developers
1•cyndunlop•15m ago•1 comments

And no more Copyleft, either

https://davegriffith.substack.com/p/the-one-thing-that-ai-generated-code
2•dxs•15m ago•0 comments

Computers Don't Argue (1965) [pdf]

https://nob.cs.ucdavis.edu/classes/ecs153-2019-04/readings/computers.pdf
1•gfitz•15m ago•0 comments

Ask HN: What is your way to go for serious iOS bugs?

1•Lausbert•16m ago•0 comments

Android, Epic, and what's behind Google's 'existential' threat to F-Droid

https://news.slashdot.org/story/26/03/16/0255231/android-epic-and-whats-really-behind-googles-exi...
2•MilnerRoute•17m ago•0 comments

Abusing Customizable Selects

https://css-tricks.com/abusing-customizable-selects/
1•speckx•17m ago•0 comments

Leadership Begins with Trust

https://mcleanonline.medium.com/set-your-people-free-bd112b2da904
2•sabinews•18m ago•0 comments

Federal Reserve Maintains Rates

https://www.federalreserve.gov/newsevents/pressreleases/monetary20260318a.htm
2•zelias•21m ago•0 comments

Why your brain has to work harder in an open-plan office than private offices

https://theconversation.com/why-your-brain-has-to-work-harder-in-an-open-plan-office-than-private...
2•PaulHoule•21m ago•0 comments

US Military confirms use of 'advanced AI tools' in war against Iran

https://www.aljazeera.com/news/2026/3/11/us-military-confirms-use-of-advanced-ai-tools-in-war-aga...
4•_____k•21m ago•0 comments

AI firm Anthropic seeks weapons expert to stop users from 'misuse'

https://www.bbc.co.uk/news/articles/c74721xyd1wo
1•_____k•22m ago•0 comments

Security Teams Waste 43% of Response Time on Manual Context Gathering

https://www.upguard.com/press/new-upguard-research-security-teams-waste-43-of-response-time-on-ma...
1•upguardnews•23m ago•0 comments

Show HN: Store and reuse your Claude Code plans

https://github.com/ChernovAndrey/Planectra
1•ChernovAndrei•24m ago•0 comments

Why Lab Coats Turned White

https://www.asimov.press/p/lab-coat
1•mailyk•24m ago•0 comments

2025 ACM Turing Award Goes to Charles H. Bennett and Gilles Brassard

https://awards.acm.org/about/2025-turing
2•taubek•25m ago•0 comments

Show HN: Pertmux – A TUI to unify your coding agents, MRs and worktrees

https://github.com/rupert648/pertmux
1•pure-orange•26m ago•0 comments