Laso Finance gets messages like this every single day. Incredibly gratifying.

🛠️ 2:00pm | Building on Ethereum in the AI Era Hosted by @cryptomastery_ of @1clawAI

An OpenAI agent that sends email with a key it never sees. Marcus Rein wired Mozilla's any-agent and the OpenAI Agents SDK into 1Claw so your credentials never reach the model's context window. The whole thing is one decorator. @mozilla @OpenAI @Marcus_Rein_ 1claw.xyz/blog/openai-agent-…

We broke 1500 followers! 👏 Thanks for the support fam. 🙏🏼

So my theory about you being thanos was indeed correct

AGENT HOUR /013: @1clawAI by @cryptomastery_ | HSM + TEE signing across Bitcoin, Solana & XRP. @deluquant by @david_tomu | Quant intelligence across Base & Robinhood Chain. nitter.net/i/broadcasts/1AKEmmdbW…

Day 5. Final chain of the week. 1Claw now supports Cardano. All @1clawAI agents can generate Cardano wallets backed by Ed25519 keys stored in physical hardware. Most chains give you a window when a private key leaks. Ethereum drains one transaction at a time. You can rotate and limit the damage. Cardano does not work that way. A single signed transaction can consume every UTxO in a wallet simultaneously. One transaction. Entire balance. One block confirmation. Gone. Private keys never touch the agent. Never in memory. Never on disk. Never in logs. The key is generated inside the HSM and stays there. When an agent needs to send ADA, it submits an intent with a recipient and a value. The server fetches UTxOs and protocol params from Blockfrost, constructs the transaction, signs inside the HSM, and broadcasts. Guardrails are enforced at the signing proxy before the key is ever accessed: ► tx_to_allowlist: restrict recipients to known addresses ► tx_max_value: cap per-transaction ADA ► tx_daily_limit: rolling 24hr spend limit ► tx_allowed_chains: lock the agent to Cardano only A compromised agent cannot bypass these. They are not in application code. Prompt injection cannot reach them. That wraps the week. Bitcoin, Solana, XRPL, Tron, Cardano. All live. Read the docs: docs.1claw.xyz/docs/guides/i… Read the blog: 1claw.xyz/blog/cardano-inten… @InputOutputHK @CardanoStiftung @IOHK_Charles @MinswapDEX @cardano @cardanofeed @TapToolsio

🎥 going live with @cryptomastery_ to talk about @1clawAI and his general sloperation nitter.net/i/broadcasts/1pKkOOQRq…

AI is for everyone nitter.net/i/broadcasts/1YxNrrkRV…

.@surshar_ is cracked. Unit of work > tokenmaxxing... Introducing the "quirq"— test it out! Looking for contributors and researchers to experiment with. Full litepaper here: xo.builders/whitepaper

🎥 going live with @cryptomastery_ to talk about @1clawAI and his general sloperation nitter.net/i/broadcasts/1pKkOOQRq…

Sign on the dotted r-address. Day 3! 1Claw now supports XRP. All @1clawAI agents can generate XRPL wallets backed by Ed25519 keys stored in physical hardware. Private keys never touch the agent. Never in memory. Never on disk. Never in logs. The key is generated inside the HSM and stays there. When an agent needs to transact on XRPL, it submits an intent. Simple payments take a recipient, value, and optional destination tag. For anything more complex, all 31 XRPL transaction types are supported: Payments, TrustSet, EscrowCreate, OfferCreate, NFTokenMint, AMMDeposit, and more. Guardrails are enforced at the signing proxy before the key is ever accessed: ► tx_to_allowlist: restrict recipients to known r-addresses ► tx_max_value: cap per-transaction XRP ► tx_daily_limit: rolling 24hr spend limit ► tx_allowed_chains: lock the agent to XRP only ► xrpl_allowed_tx_types: lock the agent to specific transaction types. A payment agent should not be able to sign a SignerListSet or AccountDelete A compromised agent cannot bypass these. They are not in application code. Prompt injection cannot reach them. Read the docs: docs.1claw.xyz/docs/guides/i… Read the blog: 1claw.xyz/blog/xrpl-intents-… @Ripple @bgarlinghouse @XRPLedger @XRPLF @xrpscan @RippleXDev

SPIDER Agents: Chainlink and Agentic Infra nitter.net/i/broadcasts/1YxNrrPyB…

your agent can now spend in the real world. the ampersend x402 skill lets agents pay any x402 endpoint directly, with your spending limits enforced on every transaction. today we're shipping a card command built with @LasoFinance: single-load US virtual prepaid cards, issued for the exact checkout amount, ready to use at any online merchant. raw card numbers stay masked from the agent. every payment needs approval from both the agent and ampersend. your funds sit in a self-custody wallet on Base the whole time. x402 just became a first-class payment primitive for agents, and real-world checkout is the first thing it unlocks. check it out: ampersend.ai

Day 2.... Lets go! 1Claw just added @solana support. All @1clawAI agents can now generate Solana wallets backed by Ed25519 keys stored in physical hardware. Private keys never touch the agent. Never in memory. Never on disk. Never in logs. The key is generated inside the HSM and stays there. When an agent needs to send SOL or interact with an SPL token, it submits a transaction intent with a recipient and a value. The server fetches the latest blockhash, constructs the transaction, signs inside the HSM, and broadcasts. ATA derivation is handled automatically. And because not every agent should be able to send to any address or any amount, guardrails are enforced at the signing proxy before the key is ever accessed: ► tx_to_allowlist: restrict recipients to known addresses ► tx_max_value: cap per-transaction value ► tx_daily_limit: rolling 24hr spend limit ► tx_allowed_chains: lock the agent to Solana only A compromised agent cannot bypass these. They are not in application code. Prompt injection cannot reach them. Read the docs: docs.1claw.xyz/docs/guides/i… Read the blog: 1claw.xyz/blog/solana-intent… @solana @aeyakovenko @rajgokal @JupiterExchange @jito @heliuslabs