Verify and retain transaction evidence
Replay a supplied transaction sample and combine results into an inspectable Proof Set.
Define what you are checking
Coverage uses every supplied transaction as the denominator. Include transactions expected to carry the code, including omissions. Filtering a dataset only by the expected Builder Code hides the gaps. The site’s public registry contains selected snapshots, with no hosted ingestion or runtime RPC requests.
Replay the published Bao Toolkit example
This command reads a public Base mainnet transaction for Bao Toolkit’s own code. It requires network access to the public RPC. Use it as a reproducible example; do not use Bao Toolkit’s Builder Code for your own app.
pnpm exec bao replay \
--builder-code bc_vwmzy653 \
--hashes 0x6573344cfb346c886806804fb8f8b6cc510c30d7974a1a69c11452a5f8fe4926 \
--chain-id 8453 \
--rpc-url https://mainnet.base.org \
--format json \
--output bao-replay.jsonCombine reports into a Proof Set
Use reports for the same Builder Code. Bao Toolkit re-decodes calldata and recalculates statuses and counters. Conflicting evidence fails validation. A passing set requires every unique transaction to be RPC verified and attributed to the expected code.
pnpm exec bao proof-set \
--builder-code bc_vwmzy653 \
--title "Bao Toolkit example" \
--input bao-replay.json \
--output bao-proof-set.jsonChoose the public evidence
Share only artifacts you intend to make public. Public reports can include transaction hashes, public calldata, decoded codes, timestamps, network metadata, and explorer links. Retain acquisition context and scope. Keep RPC credentials, private keys, and private application data out of reports.