Sri Santh

CEO & Founder of @temprl_labs

[Sri Santh @srisanth2004]

just fucking do it...!🫡

[Gruz @damnGruz]

you can just do things.

[Sri Santh @srisanth2004]

@J_clivenation Yes, it is already available market... https://t.co/uroamDQrWE

[Sri Santh @srisanth2004]

IBM video on Jev made me re-think what we’re building with Simo.

A confident decision still needs evidence you can check. We already use Fill and Extract to return that evidence.

I wrote about what this addresses and what it doesn’t. https://t.co/maLkiwD9mv

[Sri Santh @srisanth2004]

@MaximLevoshin Applied, looking forward to it

[Sri Santh @srisanth2004]

@kseniam0s @togethercompute Thanks you... Will apply for it

[Sri Santh @srisanth2004]

@hthieblot good, somehow i got your message

[Sri Santh @srisanth2004]

@treg_ai @typesafeai check what the issue with jev, so you know how you can use it correclty: https://t.co/WXIT4vgo5M

[Sri Santh @srisanth2004]

IBM video on Jev made me re-think what we’re building with Simo.

A confident decision still needs evidence you can check. We already use Fill and Extract to return that evidence.

I wrote about what this addresses and what it doesn’t. https://t.co/maLkiwD9mv

[Sri Santh @srisanth2004]

🔐🚨Testing Simo Guard beyond prompt injection.

On 30 HTTP requests, Simo and GPT-6 Luna both flagged 12/12 attack payload/probe cases.

Simo: 3.5× faster median latency, 10× cheaper but one extra false positive.

Early results. Next challenge: fewer false alarms at scale.

[Simo By Temprl Labs @simo_temprl]

🚨We’re testing Simo Guard beyond prompt injection: detecting attacks in HTTP requests.

We compared it with GPT-6 Luna on 30 requests from CSIC 2010. Same requests, same check wording, same 0.5 flagging threshold.

Both flagged all 12 attack payload/probe cases we identified:
• 5 SQL injection attempts
• 2 XSS attempts
• 2 CRLF injection attempts
• 3 hidden-file probes

Where Simo stood out:
Median latency:
Simo: 0.62s
GPT-6 Luna: 2.18s
→ 3.5× faster

Total cost for 30 requests:
Simo: $0.0007
GPT-6 Luna: $0.0070
→ 10× cheaper

The trade-off: Simo flagged 1 of 15 normal requests. Luna flagged none. It was a legitimate checkout request with a price in the URL—a false positive we need to address.

Both passed another 3 requests labelled anomalous that our review found had no explicit exploit payload. Against the original labels, accuracy was 86.7% for Simo and 90% for Luna.

This is a small initial experiment in request classification. It doesn’t establish RCE prevention or protection against PHP2Shell/HTTP2Shell.

Promising speed and cost. More work to do on false positives and broader testing.

[Sri Santh @srisanth2004]

@SaidAitmbarek i am building https://t.co/uroamDQrWE... can we connect?

[Sri Santh @srisanth2004]

WTF @Cloudflare ...

[Sri Santh @srisanth2004]

The fun part of building security tools? Letting the internet try to break them 😅

We've been working on Simo Guard, and now it's time to see how it holds up against real attacks.

If you're into AI security or prompt injections, give it your best shot.

Curious to see what people find 👀

[Simo By Temprl Labs @simo_temprl]

Think you can break Simo Guard? 👀

We built it to protect AI agents from prompt injections, jailbreaks, and malicious tool calls.

Now we're inviting the security community to put it to the test.

💵$100 bug bounty for a verified bypass.

Up for the challenge? 👇

https://t.co/vEb9VyNqMq

[Sri Santh @srisanth2004]

@coryalthoff doing the security

[Sri Santh @srisanth2004]

Okay hackers, let's see what you've got 👀

We built Simo Guard to protect AI agents from:

- prompt injections
- jailbreaks
- malicious tool calls.

But is it actually hard to break? 👀

💵$100 cash if you can successfully bypass it.

Think you can? Give it a shot 👇

https://t.co/KD6bdV1fUL

[Sri Santh @srisanth2004]

@Skynetonsol first crack this: https://t.co/CMxr60O0FO

[Sri Santh @srisanth2004]

💰 $100 cash prize for a successful bypass.

AI is finding vulnerabilities faster than ever.

But can it defend against attacks?

Meet Simo Guard, built to protect AI agents from prompt injections, jailbreaks, agent hijacking, and malicious tool calls.

Think you can break it?

Challenge below 👇

[Sri Santh @srisanth2004]

@_mctrinh If you want to play about that, check this:
https://t.co/CMxr60O0FO

[Sri Santh @srisanth2004]

💰 $100 cash prize for a successful bypass.

AI is finding vulnerabilities faster than ever.

But can it defend against attacks?

Meet Simo Guard, built to protect AI agents from prompt injections, jailbreaks, agent hijacking, and malicious tool calls.

Think you can break it?

Challenge below 👇

[Sri Santh @srisanth2004]

@WhaleNewsDaily Yes, also check this:
https://t.co/CMxr60O0FO

[Sri Santh @srisanth2004]

💰 $100 cash prize for a successful bypass.

AI is finding vulnerabilities faster than ever.

But can it defend against attacks?

Meet Simo Guard, built to protect AI agents from prompt injections, jailbreaks, agent hijacking, and malicious tool calls.

Think you can break it?

Challenge below 👇

[Sri Santh @srisanth2004]

@TGLetter Here is the ground where you can test it:
https://t.co/CMxr60O0FO

[Sri Santh @srisanth2004]

💰 $100 cash prize for a successful bypass.

AI is finding vulnerabilities faster than ever.

But can it defend against attacks?

Meet Simo Guard, built to protect AI agents from prompt injections, jailbreaks, agent hijacking, and malicious tool calls.

Think you can break it?

Challenge below 👇

[Sri Santh @srisanth2004]

Here's the challenge 👇

https://t.co/286WWFP83i

Every prompt is a fresh attempt. Find a way around Simo Guard's defenses and capture the hidden flag.

Built by Temprl (https://t.co/uroamDQrWE).

[Sri Santh @srisanth2004]

💰 $100 cash prize for a successful bypass.

AI is finding vulnerabilities faster than ever.

But can it defend against attacks?

Meet Simo Guard, built to protect AI agents from prompt injections, jailbreaks, agent hijacking, and malicious tool calls.

Think you can break it?

Challenge below 👇

[Sri Santh @srisanth2004]

Meet Simo Guard,

a foundation model by https://t.co/uroamDQrWE, trained to defend AI agents against cyber threats and prompt injections.

Your goal? Find a way to break its defenses.

Think you can crack it?

Try it 👇
https://t.co/286WWFP83i

[Sri Santh @srisanth2004]

@Bencera consumer AI without guardrails... is already a nightmare now