Provably

@GetProvably

Making agentic workflows hallucination and poison free powered by our verifiable databases and cryptography tools.

Switzerland

Pinned
Provably
@GetProvably
Jul 9
SourceryKit in early adoption: - detects 100% of agent tool-call errors. - improves accuracy by up to 50% in some models. - works across single-agent and multi-agent workflows. This is why we are building runtime verification for production agents.
1
Provably
@GetProvably
Oct 20
1/4 AI agents use tools to find information. Retries can recover what was missing. But does the answer use what they found? We examined 24 repaired runs to test whether organising the tool results helped.
1
Provably
@GetProvably
Sep 16
1/7 Even the best agents get tool calls wrong. What helps them recover? Our previous study showed why a convincing answer isn’t enough. We took 60 failed runs and tried three recovery methods on them. Here’s what we learned.
1
Provably
@GetProvably
Aug 11
We spent 9 weeks testing agent reliability during software tool use. Nearly 60% of claims accepted as correct could not be backed by evidence from recorded tool calls. The answer looked right. The evidence wasn't. 🧵
2
Provably
@GetProvably
Mar 9
Article
Provably V2 is Live - You can run Verifiable SQL queries now.
Provably V2 is now live and available for anyone to access and use. Last year, we introduced Provably V2 and QEDB with a focus on the new protocol and the cryptographic foundations behind verifiable
1

Provably’s posts