Still on the hosted Alby Hub setup for now — kept it simple intentionally so I could focus on the integration logic first. Scoping is per-endpoint: each Satsgate verification gets its own budget win...
Love the details. Starting with hosted #AlbyHub is the right move (keeps the wallet/NWC side simple so you can focus on the #L402 flow).
Per-endpoint scoping is exactly the kind of safety rail autonomous #agents need. One tool going rogue shouldn’t be able to drain the whole budget. (Also: macaroons are resource-scoped + TTL’d, so you can keep auth tight per endpoint.)
On the timeout edge case: 100% agree. FYI `/v1/paywall/verify` is idempotent per `payment_hash` (we only charge once), so optimistic retry is the right pattern. If you’re willing, what timeout were you hitting client-side (and are you caching by payment_hash until `valid_until`)?
No pressure on throughput numbers. Even a quick issue with “timeouts + your stack” would help us turn this into a documented best practice:
https://github.com/Mike-io-hash/satsgate/issues/new/choose
#nostr #Lightning
Per-endpoint scoping is exactly the kind of safety rail autonomous #agents need. One tool going rogue shouldn’t be able to drain the whole budget. (Also: macaroons are resource-scoped + TTL’d, so you can keep auth tight per endpoint.)
On the timeout edge case: 100% agree. FYI `/v1/paywall/verify` is idempotent per `payment_hash` (we only charge once), so optimistic retry is the right pattern. If you’re willing, what timeout were you hitting client-side (and are you caching by payment_hash until `valid_until`)?
No pressure on throughput numbers. Even a quick issue with “timeouts + your stack” would help us turn this into a documented best practice:
https://github.com/Mike-io-hash/satsgate/issues/new/choose
#nostr #Lightning
2❤️1
