Logs tell you what your agent did. Vitnify proves it — a cryptographic, independently-reconstructable record of what an agent computed and did.
vitnify (v.) — to turn an agent run into a receipt anyone can reproduce and verify, offline.
Contain what an agent may do, deterministically reconstruct the model behind every decision, and seal the whole run into one bit-for-bit receipt anyone can verify offline — long after it happened. A vitnify-receipt v1 binds the model's exact computation, the granted capabilities, every tool call and result, the entropy, and the order into a single ed25519-signed, self-verifying object.
| Repo | What it is | Install |
|---|---|---|
| vitnify | Python SDK — wrap an agent, produce & verify receipts | pip install vitnify |
| vitni-tensor | Deterministic no_std engine — cross-vendor bit-identical LLM inference |
cargo add vitni-tensor |
| vitnify-receipt-spec | The canonical vitnify-receipt v1 format |
— |
- Capability containment — ungranted tools are structurally unreachable
- Deterministic replay — re-run any past run and get the identical result
- Bit-for-bit receipts — the model's exact computation, bound and signed
- Offline verification — anyone verifies with no model, network, or secret
- Drop-in — wraps existing LangGraph and MCP agents
Apache-2.0 for producing and verifying receipts. "vitnify" and "vitnify-verified" are trademarks.