Repository navigation
Receipts for Franklin's wallet-backed agent work? #419
hheskihoran
started this conversation in
Ideas
Replies: 1 comment 1 reply
|
For your per-request versus job-level question, I’d put the completion claim at the job boundary and have that receipt reference request receipts as supporting evidence. That’s a proposed evidence model, not a claim about where ClawRouter can emit them. A receipt after each paid call could document the individual interaction, but I wouldn’t let “this call was paid” stand in for “the agent completed the job.” Payment belongs in settlement status; delivery needs a separate success claim, and signatures alone don’t independently verify that claim. The job receipt would be where a counterparty looks for the claimed outcome, with the request receipts backing it rather than substituting for it. |
1 reply
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Uh oh!
There was an error while loading. Please reload this page.
BlockRun's repo describes Franklin as an agent with a wallet, and ClawRouter runs on the user's side. That raises a design question: once a routed request is paid for, what independent evidence shows the agent's work happened?
I maintain INAM Protocol. It defines a two-signature Execution Receipt, an optional independent verifier verdict, and a transparency log. It never holds or moves funds. One idea is for ClawRouter to emit a receipt after each paid call, so a counterparty can check the history before it hires an agent.
Would per-request receipts fit your router's design, or would you attach them at the job level?
https://github.com/inamprotocol
All reactions