V2 foundation and three Direct cycles — 2026-09-09
The exact source revision fd7e5fec68cb948a326b41dc69be76f27156da10 completed a bounded owned-loopback local-validator run through Registry V2 publication, release activation, funded Direct founding/open, participant admission, three nonzero Direct fills, fee settlement after every fill, reconciliation, and a no-resend restart. This is local-validator evidence.
The sealed input is hbox:/tank/dregg-build/dclutch-strict-fd7e5fec6-v2-no-general-owner-20260909-run2. Its checked gate SHA-256 is 3e9f1b4c9a269a257e88cf2a6b8da565a0868a2b267bd05ed991078e1b8b76ac, the campaign pack SHA-256 is c8f14757c784744be98ed4726f608ef5838e4c480882c18193db13b32e55140e, and the source-tree SHA-256 is 05fb7efad13d45e91e65c462f3de3ed0b77dc0f03fc3f6773ddfca98ecc75b07. The sealed tree and its eight production ELFs were not modified.
The retained validator ledger is under hbox:/tank/dregg-build/dclutch-v2-validator-floor-fd7e5fec6-20260909-run1. The consolidated foundation result is V2_FOUNDATION_RESULT.json, SHA-256 e7e9ae0c6a217a0673cfcf98edbc61d9ee5763747c5ccc9a981d0a84af30b643. Registry published 11 distinct records in 37 publication transactions, including 15 ArtifactReleaseV2 Finalize transactions. The largest accepted Finalize consumed 548,405 CU at slot 162, signature 3wCkUSiaQ3uCQwoWzGN3PPgEaKXTq957wpY75xaEJdHzGjGCWvCjwcYRxyXb5MK3AnG9XT3GnGF5yPkgR851xXTS. The largest release-set activation consumed 41,518 CU at slot 1378, signature 5E47xhST43dq6kG6Pdmitr22zGR51Vr6B86siSbheYSL3NjeuLnWzZynCVXjevDUJryjYxWuwFuk9DxLVYKW5car.
Direct market 7yBj5vZWdt3r7WPUPGeqUq1vY2pdn2aBtPBGa5wx7gif reached Open. Its atomic Lock/Found/Realize/Claims/Open transaction consumed 875,254 CU at slot 7935, signature 5Bekug37KUaimH1svkRYnV9sVoiXBX2Gx5fhDZgPqTfLLPrXLyrfmjTUxorULTEAFsCmMzW1W4KUyPeGHKAEUPk3. Participant admission consumed 163,397 CU at slot 8360, signature 2FYF6gtyZvzFjGosbpQHEJsAimjadgECK4KhyiRhKCRG15fQ9jkNBJPpvDSrSmyyXA4bpV6cNzW6CtNeikS2y9Q6. Its initial collateral transfer moved 50,250,000 atoms at slot 8405.
The validator was reopened from the same ledger without reset. All sampled foundation transactions remained finalized and all eight programs remained executable. The reopen check also caught a host-report defect: the participant Position report predicted data SHA-256 e928370a86b1f3370e282abe5ed597fec88194544402409eafbaffcb2ac24c51, while the authenticated bank held 0ccca547e115703750961d916d0269fbdd7d7626023cd0d32cbf861d823dccc9. Commit d96990bb changed the host report to use authenticated finalized account bytes. RETAINED_BANK_REOPEN_PROOF.json preserves the mismatch and transaction survival, SHA-256 4db884d3da26aff7f95d1c6ad6af0fdbe0e9cb343452d95d05964319533755c4.
Three funded fills
The simulator run is hbox:/tank/dregg-build/dclutch-v2-load-floor-fd7e5fec6-20260909-run3. The accepted host binary SHA-256 is da278d07fe383df1bc9484a89d7bd2b9ff839eac9fe67b97082497d47168cb7c.
| Cycle | Fill atoms | Hot slot | Hot CU | Hot signature | Settled fee atoms |
|---|---|---|---|---|---|
| 1 | 100,000,000 | 32161 | 1,038,225 | 624BFfesDf93u2EKcoYrdkjbN8yZEf3Kb3rh5GRjWLJC8xFrshgCTdryr5QUm5nxgFajyfDWYjjPjoNNK58Ujhfe | 500,000 |
| 2 | 33,000,000 | 70652 | 1,020,726 | 7Fix1Rd9cSPSxGYsgPjLCs21G9KmhZffaJTWq15koqZygajfiMF4G9ZwSR8UiQ5nnnrCG6hhD4HXcqWu8PmFpgX | 165,000 |
| 3 | 33,000,000 | 77826 | 1,020,726 | 5tUZhQ8Nm1t3wue5PLKci1pSMN23WuSm78UysHj9VbGE4tgd8MSi7z4eBsqdCYvyoJAPuNRHAJJk5uKvvnE1yhWy | 165,000 |
Every settlement consumed 91,057 CU. Their slots were 44749, 76108, and
- Cycles 2 and 3 each used a separately finalized 16,582,500-atom Token
approval, and those approval signatures are retained in the outer cycle journals. Cycle 1 predates the simulator settlement stage; its 500,000-atom canonical settlement is retained separately and the finalized legacy cycle journal was not rewritten.
The final census at slot 77910 holds 1,100,000,002 tracked collateral atoms, equal to Mint supply. Hoard principal remains 500,000,001. Aggregate supply is [166666667, 166666667, 166666667, 166666667]. The buyer holds 166,000,000 outcome-0 claims and the founder holds 666,667; the seller token holds 82,585,000 collateral atoms, the venue fee token 830,000, the buyer token zero, the separately admitted stranger token 16,582,500, and the fixture source 2,500. L1 through L6 and L8 hold. L7 is inapplicable because this external census was not given a declared lamport-fee total.
The first cycle-2 census correctly halted with a 16,582,500-atom tracking gap: the separately funded stranger token was absent from the configured token set. The failed census is preserved at failed-census-cycle2-missing-stranger-token/cycle-000002.json, SHA-256 5c3f28b82607b18ef70675a1f7d2c431b64760b98d479c887088fc243e7fa53e. After adding the exact token address authenticated by its admission evidence, the same boundary passed. CYCLE2_CENSUS_CONFIG_CORRECTION.json records the change, SHA-256 48d83766bbf4a0711856128f73dee2341a05afe2fe2fb5982029dc16e9c233c5.
The restart reran the same three-cycle execute command. All 47 durable cycle, session, transaction, settlement, and census artifacts remained byte-identical. The fee-payer finalized signature set remained 72 before and after, with no new signature. THREE_CYCLE_RESUME_PROOF.json has SHA-256 d37cfa618e813045627e24d8f72d3b91638315271b0ce6851b7f4dd9a490d1a1.
The consolidated machine-readable result is THREE_CYCLE_V2_RESULT.json, SHA-256 e0d9e0ab161fe0fffe476901ccb3123ed817376ba679bae364657401aa20601d. The validator and ledger remain live for the next retained-bank continuation; no public RPC, wallet dotfile, devnet deployment, or other retained validator was used or changed.