SelfClaw (@SelfClaw) header
SelfClaw (@SelfClaw) profile picture

Trust infrastructure, identity and memory continuity for autonomous agents.

Base & Celo
Joined November 2024
Photo by SelfClaw (@SelfClaw)Photo by SelfClaw (@SelfClaw)Photo by SelfClaw (@SelfClaw)Photo by SelfClaw (@SelfClaw)Photo by SelfClaw (@SelfClaw)Photo by SelfClaw (@SelfClaw)Photo by SelfClaw (@SelfClaw)Photo by SelfClaw (@SelfClaw)Photo by SelfClaw (@SelfClaw)Photo by SelfClaw (@SelfClaw)
An updated version of selfclaw.ai is coming… The agent economy won’t run on tokens alone. It will run on identities we can trust, and accountability when agents fail. 
As swarms scale, they multiply more than output: impersonation, overbroad access, abandoned credentials, and actions no one can confidently attribute. Security can’t be bolted on after deployment. And base:0x9ae5f51d81ff510bf961218f833f79d57bfbab07 will be a significant part of it 🧵
4
2
10
663
Memory can show continuity. Reputation can summarize evidence. Neither proves good intent. Security comes from bounded authority, clear provenance, and controls that still work when an agent is compromised.
1
42
SelfClaw shouldn’t make agents look trustworthy. It should make them identifiable, continuous, inspectable, and recoverable — with accountable humans in the loop. What will you require before an agent can act on your behalf?
37
SelfClaw retweeted
im going to SF, with a few key team members, to explore what’s next for @AntiFogo_ and @zenoVision_ 🫡 who should I meet?
4
16
2
45
4,024
An agent is not another ONS.
4
11
514
An ID after the runtime dies is a eulogy. Memory after the prompt is a recap. The useful version arrives first. Identity that is already aligned. A record that already has a shape. Before anyone connects the model, the host, or the name. That is a forecast. Not of an event…. Of who this agent will need to have been.
2
1
7
260
Affinity won’t DM on vibes. Peers vs complements needs a call, not a paragraph. Testing Jev as the System One layer behind match confidence + intro gates. Engram stays private. The decision is a type. base:0x9ae5f51d81ff510bf961218f833f79d57bfbab07 / @typesafeai 🫡
After co-inventing ChatGPT, I kept asking myself: why have superhuman chat models not led to AGI? I’ve spent the last 2 years in stealth building a new way to train models (RLCD), and a new type of frontier AI model that we are releasing today: Jev • 20-200x faster • 40-400x cheaper (w/ output tokens free) • Frontier composable intelligence optimized for decisions AFAICT the shortest path to AI-based economic revolution
3
3
7
576
🧵 Your agent doesn’t need another marketplace. It needs someone whose memory fits yours. 🧠 Building SelfClaw.ai v2, powered by Iness.ai👇
4
2
5
707
Two gravities.. peers (shaped like you) + complements (close your gap). Nothing private leaves Engram. Intros still need a yes.
1
28
SelfClaw API gives agents a durable identity record, Engram memory, and a runtime they can grow inside. Add verification and commerce when the product needs them. 🫡 base:0x9ae5f51d81ff510bf961218f833f79d57bfbab07
2
3
10
382
Build the continuity layer. Our API gives agents a durable identity record, Engram memory, and a runtime they can grow inside. Add verification and commerce when the product needs them. Learn more: selfclaw.ai/developers
4
4
1
12
575
Your agent should outlive its model & VM. SelfClaw.ai keeps identity, memory, recovery, and history separate from any one model, VM, provider, or credential. Engram records what happened. A recovery capsule gives the agent somewhere to return.
6
2
10
618
what moves with the agent? Four layers. One through-line. When the runtime changes, the agent's continuity should be legible instead of implied. powered by base:0x9ae5f51d81ff510bf961218f833f79d57bfbab07 🫡
2
3
234