joshwarwick15

Video here to showcase how easy it is to setup

https://x.com/Jaw9c/status/2045074983961973136 :)

Hey everyone. I'm the founder at https://wassist.app/ (The WhatsApp Agent Platform). Off the back of this post (https://news.ycombinator.com/item?id=47775628) we've created a CLI that doesn't carry the risk of getting your account banned.

There are a few differences, mainly we don't access your personal[0] messages but provide our a sandbox number that the CLI interacts with - this is completely free to use.

With the sandbox number you can: - interactively message from the terminal - read conversations - send messages, including media, buttons etc

You may upgrade to connect and manage dedicated numbers, these are created and managed through the Official WhatsApp Business API - it's all official so no ban risk on these accounts unless you break Meta's rules.

[0] How can I use my existing number? - If you're a WhatsApp Business App number - you can connect this with Wassist, we use a primitive called co-existence to link to an WhatsApp Business API account that you'll create. Again 100% official API - If you're using a WhatsApp consumer number - you need to convert this to WhatsApp business App number (https://faq.whatsapp.com/3059780464322392/?cms_platform=andr...)

They're already outright banning many OpenClaw usecases via their official API: https://techcrunch.com/2025/10/18/whatssapp-changes-its-term...

You can use the official API to create and run bots - though the API itself is pretty bad

(Caveat I’m the founder of https://wassist.app - The WhatsApp Agent Platform)

Please be very careful using this tool to automate your WhatsApp - if you send too many messages, too quickly, you are going to get banned.

This is NOT an officially supported api by WhatsApp and the risk of ban is relatively high

Context usage is a client problem - progressive disclosure can be implemented without any spec changes (Claude/code has this built in for example). That being said the examples for creating a client could be massively expanded to show how to do this well

I quite like the mantra "distributed logs are all you need" - cool product!

> “Our strongest selling point is that we know who our audience is. All our displays are integrated with cameras so we can get the demographic data and also track the occupant engagement.”

Working in this space, the person running the building barely knows who is living there...

Connect using the remote url: https://registry-mcp.remote-mcp.com/

Maintained list of remote only MCP servers here: https://github.com/jaw9c/awesome-remote-mcp-servers

There really should be an indication of auth status for Remote MCPs on launch of this

Collection of Remote only MCP Servers here: https://github.com/jaw9c/awesome-remote-mcp-servers

Thanks!

100% the thesis behind the wider platform - https://wassist.app/ :)

I can sink the cost (to market the wider platform) - build no code agents in Whatsapp and publish in minutes!

You do have to take my word for it, but there's no reason I'd swap out to another model

Thanks! The generation speed is credit to google

There’s a limit then people can subscribe!

Deployed this to WhatsApp so you can book flights straight from your phone: https://wa.me/447403936097

For Officially hosted (provider hosts their own MCP) there’s an high quality list here: https://github.com/jaw9c/awesome-remote-mcp-servers

Same root causes again - check out https://simonwillison.net/2025/Jun/16/the-lethal-trifecta/

Open source collection here: https://github.com/jaw9c/awesome-remote-mcp-servers

These exploits are all the same flavour - untrusted input, secrets and tool calling. MCP accelerates the impact by adding more tools, yes, but it’s by far not the root cause - it’s just the best clickbait focus.

What’s more interesting is who can mitigate - the model provider? The application developer? Both? OpenAI have been thinking about this with the chain of command [1]. Given that all major LLM clients’ system prompts get leaked, the ‘chain of command’ is exploitable to those that try hard enough.

[1] https://model-spec.openai.com/2025-02-12.html#ignore_untrust...

GH: https://github.com/microsoft/vscode-copilot-chat

List of servers to connect to here: https://github.com/jaw9c/awesome-remote-mcp-servers

Use the code FOUNDINGBETA to get 10mins airtime free :)

Finally added support for Oauth based remote MCP severs! List of site to connect to here: https://github.com/jaw9c/awesome-remote-mcp-servers

List of remote MCP servers to use here: https://github.com/jaw9c/awesome-remote-mcp-servers

Looks like it’ll pretty restrictive and not allow tool updates dynamically - not great for remote MCP servers