02 / Future work

Do not trust the result. Enter through an explicit evidence path.

The present artifact report stands on its frozen evidence. These experiments test what remains open: full-node closure, corpus breadth, causal effects, and external reproducibility.

Close the full-node gate

Start from empty local state, reach testnet4 tip while validating every connected block, then demonstrate meaningful maintenance and the remaining live-node capabilities.

python3 Project/scripts/report.py --db Project/project.db --section consensus-runway

Run an external replication

Reproduce selected validators and the evidence pipeline without sharing the original developer’s tacit knowledge or admission decisions.

python3 Project/scripts/preflight_port_baseline.py --db Project/project.db --port <port> --strict

Control the substrate

Hold the task, environment, human policy, and agent configuration fixed while comparing a populated verification substrate with an otherwise equivalent empty one.

Protocol must be designed and preregistered before execution

Expand the negative corpus

Valid-chain replay catches false rejection. New provenanced must-reject classes and mutation-sensitive checks target silent false acceptance.

python3 Project/scripts/preflight_consensus_runway.py --db Project/project.db --port <port> --stage corpus --strict

Add a language

Use the shared ledger, fixtures, and contracts while implementing a separate consensus decision path. Do not treat the result as an independent experimental replication.