Head to head
MCP vs ONCE — Exactly-Once Reliability for AI Agent Actions
Same numbers we show everywhere else, lined up. Ratings come from users, the grade comes from our scanner, adoption comes from public registries.
ONCE — Exactly-Once Reliability for AI Agent Actions
Bio.github.GSterlingPress/onceNo reviews yet
Prevent duplicate AI-agent side effects with idempotency, verification, and durable receipts.
Full listingThe short version
Neither has been reviewed yet, so the comparison rests on the scan and the public usage numbers. The safety scan favours MCP (99/100 against 79). MCP has noticeably more public adoption.
| MCP | ONCE — Exactly-Once Reliability for AI Agent Actions | |
|---|---|---|
| User rating | No reviews yet | No reviews yet |
| Safety grade | A99/100 | B79/100 |
| Adoption | Established | Growing |
| GitHub stars | 533 | 0 |
| Downloads / week | 297 | – |
| Runs where | Hosted (remote) | Hosted (remote) |
| Gateway-ready | Yes | Yes |
| Tools | — | 2 |
| Write actions | — | 1 |
| Price per call | Free | Free |
| Licence | Apache-2.0 | — |
| Last release | 2026-09-19 | 2026-09-13 |
| Category | AI and agents | AI and agents |