Miyamura80

4 ms·

1.
by Miyamura80 10d ago
Agree! CLIs are still superior for piping etc, but as an auth middleware where the LLM doesn't see the credentials, I think it is really shaping up well
2.
Show HN: Sleeper Agents in Robot Dogs and Kinetic Prompt Injections(eito.substack.com)
4 points by Miyamura80 25d ago|0 comments
3.
by Miyamura80 26d ago
Awesome!
4.
Stdio MCP Tunnels: Connecting Mobile/Cloud Agents to iMsg/WhatsApp/Linkedin(twitter.com)
4 points by Miyamura80 1mo ago|3 comments
5.
by Miyamura80 1mo ago
Many members of our team were users of Beeper, a "unified inbox" for chat networks like WhatsApp, Telegram, iMessage, Slack, LinkedIn DMs, etc, and wanted to expose this to our agents. However, we quickly ran into the issue where
6.
by Miyamura80 1mo ago
This makes a tonne of sense. NVIDIA actively does not want a world where they only have two customers between Anthropic and OpenAI, and empowering open source seems like the most beneficial way for them to commoditize OpenAI and Anthropic,
7.
by Miyamura80 1mo ago
This is cool, we've built similar things, I'm a bit uneasy with crypto stuff involved, especially with algorand But FIDO2 passkey signature I absolutely agree should be the way to go. Just need more open standard without tokens in
8.
by Miyamura80 1mo ago
Afaik the other main issue with github downtime is the choice they made a while back coming back to bite them
9.
by Miyamura80 1mo ago
Personally never really understood A2A, it seems too complicated for ape adoption, and I do feel like MCP Spec will move fast enough to handle these adversarial situation highlighted in the blog post
10.
by Miyamura80 1mo ago
Hermes at 22 seems criminally low
11.
Jailbreaking Clawdbot to Plant Malware(twitter.com)
1 points by Miyamura80 8mo ago|1 comments
12.
by Miyamura80 8mo ago
A prompt injection attack via GitHub issues that is close to invisible, even to experienced engineers, and visible to LLMs. Clawdbot is full of security holes and we're having fun. What are some workflows you / you've seen ot
13.
by Miyamura80 9mo ago
Good point around the markdown image as an untrusted vector. Lethal trifecta is determnistically preventable, it really should be addressed wider in the indutry
14.
by Miyamura80 9mo ago
Totally agree, unfettered access to databases are dangerous There are ways to reduce injection risk since LLMs are stateless and thus you can monitor the origination and the trustworthiness of the context that enters the LLM and then decide
15.
by Miyamura80 9mo ago
In case there is an interest for solution to this, check out open.edison.watch and the accompanying blog posts on the README there.
16.
by Miyamura80 10mo ago
Sorry not familiar with this. Can you please link me?
17.
by Miyamura80 10mo ago
You actually can protect against it, by tracking context entering/leaving the LLM, as long as its wrapped in a MCP gateway with trifecta blocker. We've implemented this in open.edison.watch
18.
by Miyamura80 2y ago
Very cool, what is the tok/s on this?