feat(benchmark): add strict-v4 publishable comparison pipeline by 777genius · Pull Request #52 · 777genius/infinity-context

777genius · 2026-08-10T19:43:29Z

Что готово

strict-v4 authority lifecycle и fail-closed writer fence
durable provider-free preparation и publishable comparison pipeline
изолированный Mem0 runtime с authenticated relay и exact attestation
bounded verification актуального Codex executable с обязательным SHA-256

Проверки

provider-free full LoCoMo preparation: 1 passed
strict-v4 role/fence E2E: 2 passed
focused runtime/operator gate: 50 passed
Ruff, format и diff checks: passed

До Ready for review

SHA-bound Docker create/reopen gate
один live LLM canary
полный 2040-pair Infinity Context vs Mem0 прогон и publishable report

PR оставлен draft, пока live acceptance не завершен.

WIP: deployment acceptance remains blocked by fleet test reconciliation and a fresh isolated runtime gate.

Bind the production runner to the authenticated relay, align staging and attestation with the six-service v2 topology, and preserve immutable acceptance evidence.

Propagate the exact create or reopen environment to every Compose command so isolated runtime inspection can attest the deployed lane.

Accept the reviewed 298 MB Codex binary under a finite 320 MiB verification budget while preserving exact SHA-256 matching and bounded streaming reads.

coderabbitai · 2026-08-10T19:43:40Z

Important

Review skipped

Draft detected.

Please check the settings in the CodeRabbit UI or the .coderabbit.yaml file in this repository. To trigger a single review, invoke the @coderabbitai review command.

⚙️ Run configuration

Configuration used: Organization UI

Review profile: CHILL

Plan: Pro Plus

Run ID: 5d4461ba-ec6d-441a-8fd1-f01865d46428

You can disable this status message by setting the reviews.review_status to false in the CodeRabbit configuration file.

Use the checkbox below for a quick retry:

🔍 Trigger review

Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out.

❤️ Share

_{Comment @coderabbitai help to get the list of available commands.}

777genius added 25 commits August 8, 2026 23:23

fix(benchmark): attest v5 runtime before paid readiness

14ed225

chore(benchmark): refresh v5 runtime attestation authority

1f13a3a

fix(benchmark): preserve official v5 source whitespace

8de62f8

chore(benchmark): refresh v5 source whitespace authority

d7bf1ac

feat(benchmark): add crash-safe managed v5 recovery

ea30b63

chore(benchmark): refresh managed v5 recovery authority

3334fc5

test(benchmark): migrate unsealed abort e2e to v2

d8f3d75

fix(benchmark): separate runtime and extraction authority

250d1bf

fix(benchmark): separate runtime and extraction schemas

c3704ee

fix(postgres): allow exact benchmark tombstones

152a09d

fix(benchmark): verify immutable node authority

ac97bf6

feat(benchmark): add paged full-run manifest authority

c61a1d9

feat(benchmark): predeclare paired superiority policy

af1e3af

fix(benchmark): recover preregistration state safely

4decec3

fix(benchmark): guard recovery cold imports

0d657f3

feat(benchmark): add durable evaluation scheduler contracts

4a0a93b

feat(benchmark): persist durable evaluation state

5ce989b

feat(benchmark): add paged cleanup authority

5e0f4a0

feat(benchmark): add provider-safe scheduler v2

c713024

feat(benchmark): add strict v4 authority lifecycle

0b4ae0d

feat(benchmark): add durable publishable comparison pipeline

453b9f4

chore(benchmark): checkpoint isolated mem0 deployment

2a2dae2

WIP: deployment acceptance remains blocked by fleet test reconciliation and a fresh isolated runtime gate.

fix(benchmark): reconcile isolated mem0 deployment

493e8f2

Bind the production runner to the authenticated relay, align staging and attestation with the six-service v2 topology, and preserve immutable acceptance evidence.

fix(benchmark): preserve compose attestation environment

89bd23e

Propagate the exact create or reopen environment to every Compose command so isolated runtime inspection can attest the deployed lane.

fix(runtime): raise bounded Codex executable limit

0758567

Accept the reviewed 298 MB Codex binary under a finite 320 MiB verification budget while preserving exact SHA-256 matching and bounded streaming reads.

777genius added 4 commits August 10, 2026 19:51

style(benchmark): sort test imports

2e7f75b

fix(benchmark): align publishable provider identity

6dcf0eb

fix(tests): make adapter fixtures importable

3f26e3c

fix(benchmark): attest relay reachability

1c1a179

777genius added 30 commits August 11, 2026 16:14

fix(benchmark): refresh reviewed orchestration binding

5059454

fix(benchmark): bind official dataset authority

97c8efd

ci(benchmark): enforce provider-free contracts

056dc2d

style(benchmark): format CI contract runner

5722ddb

fix(ci): calibrate benchmark selection floor

35fef54

fix(benchmark): preserve retrieval reopen authority

9ec5aee

fix(ci): install root benchmark closure

e2a5c7c

fix(postgres): harden strict-v4 authority roles

f19ce03

test(postgres): restore strict-v4 migration roles

fffba04

feat(benchmark): add safe fresh 1+4 canary

60910e0

fix(benchmark): harden fresh canary activation

a2555ab

fix(benchmark): close fresh canary review gaps

0ad32f3

fix(postgres): reconcile strict-v4 draft lifecycle

49eb90c

fix(benchmark): restore input secret validation wiring

b0d85a6

test(sqlite): make private mode check umask-safe

fc0368b

fix(postgres): pin strict-v4 topology phases

779cfdc

fix(postgres): consolidate strict-v4 canonical writer

ff30f13

refactor(benchmark): use canonical writer credentials

aafa9c2

test(postgres): prove canonical strict-v4 boundaries

2b02c72

feat(postgres): provision self-host identities

1746e1b

feat(postgres): reconcile self-host runtime ACL

17637c5

feat(server): add self-host database commands

27a6274

test(selfhost): prove isolated PostgreSQL 18 startup

7423a4c

test(selfhost): split compose contract checks

005614c

test(postgres): align strict-v4 phase attestation

a7c4e9e

chore(benchmark): refresh mem0 source authority

c127441

feat(benchmarks): add isolated provider-free project gate

85e73dc

fix(benchmarks): mount server package in relay

69c79bb

fix(benchmarks): mount source manifest from pin authority

82aa8bd

fix(benchmarks): honor mapped evidence ownership

aebfaa8

Navigation Menu

feat(benchmark): add strict-v4 publishable comparison pipeline - #52

feat(benchmark): add strict-v4 publishable comparison pipeline#52
777genius wants to merge 98 commits into
mainfrom
fix/managed-v5-runtime-attestation-r1

777genius commented Aug 10, 2026

Uh oh!

coderabbitai Bot commented Aug 10, 2026 •

edited

Loading

Review skipped

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

1 participant

Conversation

777genius commented Aug 10, 2026

Что готово

Проверки

До Ready for review

Uh oh!

coderabbitai Bot commented Aug 10, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Review skipped

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

1 participant

coderabbitai Bot commented Aug 10, 2026 •

edited

Loading