Issues

• fix(bench): find a launch's live process through its threads and across an exec

  kamui/code-review-bench#107 · Filter by author kamui opened on Oct 11, 2026
  1

• fix(bench): tie a claim to its launch without comparing wall-clock stamps

  kamui/code-review-bench#105 · Filter by author kamui opened on Oct 11, 2026

• fix(bench): find why the review queue's process check intermittently reports unknown

  kamui/code-review-bench#103 · Filter by author kamui opened on Oct 10, 2026
  1

• feat(grading): make "named the cause only" repeatable so the scorecard can show it

  benefit: reliabilityenhancementpriority: P2
  kamui/code-review-bench#78 · Filter by author kamui opened on Oct 7, 2026

• feat(sandbox): add explicit unsandboxed host execution

  kamui/code-review-bench#61 · Filter by author kamui opened on Oct 6, 2026

• chore(bench): cut tasks at the last push instead of the merge

  benefit: reliabilityenhancementpriority: P2
  kamui/code-review-bench#60 · Filter by author kamui opened on Oct 6, 2026
  3

• feat(grading): let agents settle most adjudications and raise the uncertain ones to me

  benefit: reliabilitydocumentationenhancementpriority: P1
  kamui/code-review-bench#59 · Filter by author kamui opened on Oct 5, 2026
  7

• chore(bench): prepare the sandbox rollout and cohort readiness report

  benefit: reliabilitybenefit: token-costdocumentationenhancementpriority: P1status: deferred
  kamui/code-review-bench#46 · Filter by author kamui opened on Oct 3, 2026

• feat(skills): package a portable local review launcher

  benefit: reliabilitydocumentationenhancementpriority: P2
  kamui/code-review-bench#45 · Filter by author kamui opened on Oct 3, 2026

• feat(sandbox): enforce admission limits and measure laptop concurrency

  benefit: latencybenefit: reliabilityenhancementpriority: P1
  kamui/code-review-bench#44 · Filter by author kamui opened on Oct 3, 2026

• feat(bench): preserve container evidence and recover interrupted attempts

  benefit: reliabilityenhancementpriority: P1
  kamui/code-review-bench#43 · Filter by author kamui opened on Oct 3, 2026

• feat(bench): integrate execution modes and receipts across review arms

  benefit: reliabilityenhancementpriority: P1
  kamui/code-review-bench#42 · Filter by author kamui opened on Oct 3, 2026

• feat(sandbox): package pinned images and a shared review launch lifecycle

  benefit: reliabilityenhancementpriority: P1
  kamui/code-review-bench#41 · Filter by author kamui opened on Oct 3, 2026

• feat(sandbox): isolate provider credentials in a bounded broker

  benefit: reliabilitybenefit: token-costenhancementpriority: P1
  kamui/code-review-bench#40 · Filter by author kamui opened on Oct 3, 2026

• feat(sandbox): add rootless Podman doctor and macOS staging

  benefit: reliabilitydocumentationenhancementpriority: P1
  kamui/code-review-bench#39 · Filter by author kamui opened on Oct 3, 2026

• feat(sandbox): resolve CE peer launches and skill sandbox conflicts

  benefit: reliabilityenhancementpriority: P1
  kamui/code-review-bench#38 · Filter by author kamui opened on Oct 3, 2026

• feat(sandbox): verify Codex tools and child-thread confinement

  benefit: reliabilityenhancementpriority: P1
  kamui/code-review-bench#37 · Filter by author kamui opened on Oct 3, 2026

• feat(sandbox): default review execution to Podman with explicit host mode

  benefit: reliabilityenhancementepicpriority: P1
  kamui/code-review-bench#35 · Filter by author kamui opened on Oct 3, 2026
  1

• chore(grading): regrade the selected review cohort under the next rubric and verify it

  benefit: precisionbenefit: reliabilitybenefit: token-costenhancementpriority: P1
  kamui/code-review-bench#30 · Filter by author kamui opened on Oct 3, 2026
  5

• feat(scoring): implement grading and scoring v1

  benefit: actionabilitybenefit: precisionbenefit: reliabilityenhancementepicpriority: P1
  kamui/code-review-bench#24 · Filter by author kamui opened on Oct 3, 2026