What happened.

The Leios prototype release 2026w32 now constructs its voting committee from the on-chain stake distribution. It uses per-pool BLS keys carried in PoolDistr to sign votes and verify certificates against that committee.

The release removes the earlier placeholder committee and an insecure key-derivation shortcut. It completes the earlier path in which BLS keys could be registered but were not yet used for voting.

Why it matters.

This is a concrete consensus-plumbing step for the Leios prototype: voting is now stake-weighted and cryptographically signed rather than relying on the earlier placeholder path. It also gives SPOs and builders an early operational requirement to test.

A node votes only when started with `--shelley-bls-key` pointing to its registered signing key. Without that setting, it runs as a non-voting relay.

Limits and next receipt.

This is a Musashi devnet prototype update, not a mainnet hard fork or a shipped throughput gain. Committee selection remains stake-only, has a placeholder coverage target, and is not yet governance-configurable.

Earlier ledger state and blocks no longer decode after this change. Operators are expected to delete local state and resync from genesis, or sideload from IOG relays, during a respin expected within roughly one to two days. The next receipt to watch is that respin and successful exercise of the BLS-key voting path by devnet nodes.

What to watch

Watch for the Musashi respin and confirmation that nodes configured with registered BLS keys are participating in committee voting.

Sources and limits

Upstream references and independent checks

Digest dated 2026-08-10 · upstream model claude-sonnet-4-6. Direct links are matched to all 2 upstream source IDs.

  1. 1
    Prototype 2026w32Direct upstream source · db50267b564277e3092ca9e25aacc9eb91112049
  2. 2
    Release prototype-2026w32Direct upstream source · 730df09ee9f2165f68b6e90b468675fc1df7dab1

This Research brief was generated by Terra from a dated upstream research digest. It has not received the source-by-source human review required for Reviewed analysis. Material limit: The evidence supports a high-confidence devnet prototype milestone, but it does not establish mainnet deployment, a throughput improvement, or a finalized committee-selection design.