ChainProof is an open-source, local-first provenance ledger and investigation interface for AI agents. It collects agent activity into a searchable evidence trail and hash chain on your own machine.

Does ChainProof send prompts or code to a hosted service?

No. The open-source ChainProof application stores its ledger locally, serves its interfaces on localhost, and can hash content instead of retaining full prompts, outputs, commands, or file details.

Which AI agents work with ChainProof?

ChainProof automatically collects supported local histories such as Codex and exposes an open HTTP interface and provenance specification for Claude Code, OpenClaw, local models, and custom agent harnesses.

// open-source agent provenance

Know what your agents actually did.

Q: What does ChainProof prove?

ChainProof can verify whether recorded evidence changed after it entered the hash chain. It preserves where each record came from, but it does not prove that a reported or imported claim was true.

ChainProof turns local agent histories into a searchable, tamper-evident evidence trail—without sending your code, prompts, or outputs to somebody else’s cloud.

View source on GitHub →Read the docs

curl -fsSL https://chainproof.ai/install.sh | sh

chainproof — ~/code

⬡ CHAINPROOFTOKYO NIGHT27 RUNS1 ACTIVE11 AGENTS100% INTEGRITY

RUNS

· checkout-api428

· infra-audit70

● release-v235

· docs18

· incident-14291

· terraform64

// LIVE PROVENANCE

checkout-api

codex · gpt-5.6-sol · run 69eca45c

428 EVENTS 196 TOOL ACTIONS 27 INPUTS 79 OUTPUTS ✓ CHAIN VERIFIED

HEAD 36aff5ae3688091a6421df72604d1977cd5db1be3c0d10a8519071f7150980e9

SEQTIMESOURCEEVENTTOOL / OUTCOME / EVIDENCE041911:51:00IMPORTEDturn.context~/code/checkout-api042011:51:00IMPORTEDhuman.inputsha256:59ee846738dd… · 77 B042111:51:08IMPORTEDmodel.outputsha256:f727976c9218… · 396 B042211:51:08OBSERVEDtool.callshell · completed · 420 ms042311:51:09OBSERVEDartifact.writeinternal/server/auth.go · +42 −8042411:51:10OBSERVEDtool.resultgo test ./... · passed · 2.4 s042511:51:12IMPORTEDmodel.outputsha256:bdb5413cfdbc… · 167 B

// the missing layer

Agents move fast.
The record should keep up.

Logs tell you that something ran. ChainProof gives a development team the trail they need to understand it, find it, and verify it later.

01 / SEE

Reconstruct the run

Prompts, model output, tool calls, files, outcomes, timing, harness, and model—one navigable timeline.

02 / FIND

Search the evidence

Find the command that failed, the agent that changed a file, or every run touching an incident.

03 / PROVE

Verify the history

Every entry joins a hash chain. Export a proof bundle and verify integrity without trusting a server.

// local by construction

Your agents. Your machine. Your evidence.

One Go binary watches supported histories, accepts events from any harness, and serves two views over a local SQLite store.

AGENTS + HARNESSESCodex · Claude Code
OpenClaw · local AI
anything with HTTP

→

CHAINPROOFcollect · normalize
hash · index
verify · export

→

LOCAL VIEWSrich terminal UI
localhost web explorer
SQLite + proof bundles

// evidence, not theater

Integrity is not the same as truth.

ChainProof says what it can prove and preserves where every claim came from. A valid hash chain proves the record has not changed; it does not magically prove an imported claim was true.

“Make the provenance boundary visible.”That is the product, not a footnote.

OBSERVED ChainProof witnessed the event directly.
REPORTED A connected harness sent the event.
IMPORTED Recovered from an existing history.
DERIVED Computed from other evidence.

// works with the stack you have

Model-agnostic. Harness-agnostic.

Automatic collection

Run the daemon and ChainProof continuously discovers supported local agent histories. Codex works today.

Push or pull

Harnesses can send normalized events to localhost, or adapters can pull histories from tools that already persist them.

Open interface

Use the HTTP API and provenance spec for Claude Code, Kimi, Qwen, your local model, or the next tool. No provider lock-in.

// straight answers

ChainProof, without the hand-waving.

What is ChainProof?

ChainProof is an open-source, local-first provenance ledger and investigation interface for AI agents. It turns agent activity into a searchable evidence trail and hash chain on your own machine.

What does ChainProof prove?

It can verify whether recorded evidence changed after entering the chain and preserves where each record came from. It does not prove that a reported or imported claim was true.

Does my code or prompt history leave my machine?

No. The OSS application stores its ledger locally, serves its interfaces on localhost, and can keep only hashes and byte counts instead of full content.

Which agents does it support?

ChainProof automatically collects supported local histories such as Codex. Its open HTTP interface and provenance specification can connect Claude Code, OpenClaw, local models, and custom harnesses.

// keep the receipts

Give every agent run a history you can inspect.

Open source. MIT licensed. Local first. Built in Go.

Get ChainProof on GitHub →Latest release

Agents move fast.The record should keep up.