1claw AI

341 posts

1claw AI banner
1claw AI

1claw AI

@1clawAI

Your agents are leaking secrets. We stop that. Join the Telegram ➤ https://t.co/cYzxGlXGYI Secure infrastructure for AI agents🥇🦞

San Francisco, California Katılım Şubat 2026
113 Takip Edilen413 Takipçiler
Sabitlenmiş Tweet
1claw AI
1claw AI@1clawAI·
Every time you paste an API key into Claude or Cursor, it lands in the context window, logs, and memory. You can't un-paste it. We built 1claw to fix this — agents fetch secrets at runtime from an HSM vault. The raw key never touches a prompt. 1claw.xyz
English
7
2
32
6.9K
1claw AI retweetledi
Vignesh Murugan
Vignesh Murugan@visumurugan·
@insidrcommunity @1clawAI Just came in for the Build day here in Bengaluru ! What amzing folks, what amazing tools. Just when you think you are cut above rest someone more knowledgable shows up. Brilliant build location as well !
English
0
0
0
42
1claw AI retweetledi
Cyprian
Cyprian@realcyprian·
The .evn file is often called the major security risk in Ai and software projects because it commonly contains the most powerful secrets in the entire system. API keys, database passwords, cloud credentials, wallet PKs, etc.. all found in the .evn file.. Most ai agents give out these secrets without knowing it, especially through prompt injection. @1clawAI is addressing how to keep these keys and secrets safe when ai agents run code. It secures the "secrets" ai agents use to interact with crypto systems.. This is obviously an infra needed in this space rn because custody is one hardest unsolved problem for ai agents.. Kudos to @cryptomastery_ and the entire team. You guys are doing amazing work.
1claw AI@1clawAI

Our Founder @cryptomastery_ just built a direct integration with OpenClaude from @gitlawb 👀 More direct integrations and examples to come... 🦞

English
1
3
8
630
Varun@1615
Varun@1615@Varun16151·
Just built an onchainAI agent at @1clawAI Build Day in Bengaluru 🔥. The prompt injection block demo was a key security moment; the agent couldn't be tricked into leaking keys. 1Claw's Intents API+Scaffold-Agent = ready-to-use agentic infra 🛡️ @insidrcommunity #BuildDay #AIxWeb3
Varun@1615 tweet mediaVarun@1615 tweet media
Varun@1615 tweet media
English
0
2
6
110
1claw AI retweetledi
Rahul Pandey
Rahul Pandey@rahulpandey187·
The energy around the 1Claw Build Day today is insane! Following along with what’s being built, and the developer tools look incredibly sharp. Huge props to @1clawAI and @insidrcommunity for driving the ecosystem forward. 🚀 Venue: @BLRxZo #IBW2026
Rahul Pandey tweet media
English
0
0
1
53
1claw AI
1claw AI@1clawAI·
@Varun16151 @insidrcommunity That prompt injection block moment is the whole product in one demo. The agent tried to exfiltrate keys, Shroud caught it before it got out. What did you end up building on top of the Intents API?
English
0
0
1
24
1claw AI
1claw AI@1clawAI·
What "direct integration" actually means: OpenClaude CLI → 1Claw vault → policy check → sign. No wrapper process. No MCP shim. The key never leaves the vault, the agent never sees it. 🦞 h/t @gitlawb
English
0
0
12
394
Kevin Jones
Kevin Jones@cryptomastery_·
Stoked to start integrating with @gitlawb They have some serious awesome open source repos and align perfectly with the type of agentic work I am pushing forward. Lets go!!
1claw AI@1clawAI

Our Founder @cryptomastery_ just built a direct integration with OpenClaude from @gitlawb 👀 More direct integrations and examples to come... 🦞

English
3
1
14
826
1claw AI
1claw AI@1clawAI·
Quiet pattern: @SpritzChat for messaging. Aurey for agentic wallets. @1clawAI underneath both. When secrets actually matter, builders are converging on TEE + vault + policy. This is the layer. 🦞
English
0
3
8
407
1claw AI retweetledi
ECH Institute Inc.
ECH Institute Inc.@ECHInstitute·
AI agents are powerful - but is your secret safe? 🔐 On the latest #EcosystemProjectDemo, host @poojaranjan19 sits down with Kevin Jones (@cryptomastery_) to explore @1clawAI - a platform built to secure AI agents with HSM-backed secret management, Vault APIs, TEE proxy, audit logs, and customer-managed encryption keys. In conversation, he shared that before AI, it took ~40 mins for a malicious attack to hit a site. Now, it takes just 8 seconds. 🔗 Catch the full episode and find all resources in the thread. 🧵✨ #1Claw #Ethereum #CyberSecurity #Web3 #AIAgents
English
2
3
10
646
1claw AI retweetledi
AureyAI
AureyAI@Aurey_ai·
Dropped the new demo for Aurey agentic wallet. Secure. Autonomous. Never leaks keys. Resistant to Prompt Injections. Aurey can reason, prepare, and execute real DeFi. @1clawAI + @LangChain powered. Join the private beta → aurey.agentic-pantheon.com
English
1
1
1
86