›_ Claude Code Issues
Claude Code Issues / #56396

[BUG] [ultrareview] Free runs consumed on large-scope branch reviews with empty findings (2 sessions)

Resolved 💬 4 comments Opened May 5, 2026 by diek617 Closed Jun 2, 2026

Preflight Checklist

  • [x] I have searched existing issues and this hasn't been reported yet
  • [x] This is a single bug report (please file separate reports for different bugs)
  • [x] I am using the latest version of Claude Code

What's Wrong?

Bug

Same failure mode as #50029 — /ultrareview consumed free runs on large-scope branch reviews, returned empty findings in ~25 seconds (vs expected 5-10 min).

Two sessions affected (Pro/Max plan, today 06.05.2026)

Session 1 — Atlas/CraftBot Dashboard repo

  • Branch: review-full-atlas (created at initial commit 350f0cd, no cherry-pick — branch HEAD is initial-commit, default branch main is HEAD)
  • Comparison: main against review-full-atlas
  • Scope: 601 files changed, +124344 / -618 LOC
  • Findings: []
  • Duration: ~25 seconds
  • Free run consumed: 1/3 → 2/3
  • Session ID: 019eDUpc3TgSyYQbeUXYKtJq

Session 2 — VerwaltBot repo

  • Branch: review-full-verwaltbot (cherry-picked 201 commits from initial commit 597544c onto branch)
  • Comparison: review-full-verwaltbot → main
  • Scope: 731 files changed, +131107 / -37 LOC
  • Findings: []
  • Duration: ~20 seconds
  • Free run consumed: 2/3 → 3/3
  • Session ID: session_013EXdL7Du2PQgMVEpDWvzQD

Root cause hypothesis

Same as #50029: silent skip on large-scope bundle (>100k LOC), free run consumed despite no review delivered. No error message, no findings, no agent activity logs visible from CLI.

Request

Refund both free runs. Plus: clearer error surfacing when bundle exceeds analyzable size limit.

Environment

  • Plan: Claude Pro/Max
  • Date: 2026-05-06
  • Two separate repositories, same failure mode

What Should Happen?

Bug

Same failure mode as #50029 — /ultrareview consumed free runs on large-scope branch reviews, returned empty findings in ~25 seconds (vs expected 5-10 min).

Two sessions affected (Pro/Max plan, today 06.05.2026)

Session 1 — Atlas/CraftBot Dashboard repo

  • Branch: review-full-atlas (created at initial commit 350f0cd, no cherry-pick — branch HEAD is initial-commit, default branch main is HEAD)
  • Comparison: main against review-full-atlas
  • Scope: 601 files changed, +124344 / -618 LOC
  • Findings: []
  • Duration: ~25 seconds
  • Free run consumed: 1/3 → 2/3
  • Session ID: 019eDUpc3TgSyYQbeUXYKtJq

Session 2 — VerwaltBot repo

  • Branch: review-full-verwaltbot (cherry-picked 201 commits from initial commit 597544c onto branch)
  • Comparison: review-full-verwaltbot → main
  • Scope: 731 files changed, +131107 / -37 LOC
  • Findings: []
  • Duration: ~20 seconds
  • Free run consumed: 2/3 → 3/3
  • Session ID: session_013EXdL7Du2PQgMVEpDWvzQD

Root cause hypothesis

Same as #50029: silent skip on large-scope bundle (>100k LOC), free run consumed despite no review delivered. No error message, no findings, no agent activity logs visible from CLI.

Request

Refund both free runs. Plus: clearer error surfacing when bundle exceeds analyzable size limit.

Environment

  • Plan: Claude Pro/Max
  • Date: 2026-05-06
  • Two separate repositories, same failure mode

Error Messages/Logs

Steps to Reproduce

Bug

Same failure mode as #50029 — /ultrareview consumed free runs on large-scope branch reviews, returned empty findings in ~25 seconds (vs expected 5-10 min).

Two sessions affected (Pro/Max plan, today 06.05.2026)

Session 1 — Atlas/CraftBot Dashboard repo

  • Branch: review-full-atlas (created at initial commit 350f0cd, no cherry-pick — branch HEAD is initial-commit, default branch main is HEAD)
  • Comparison: main against review-full-atlas
  • Scope: 601 files changed, +124344 / -618 LOC
  • Findings: []
  • Duration: ~25 seconds
  • Free run consumed: 1/3 → 2/3
  • Session ID: 019eDUpc3TgSyYQbeUXYKtJq

Session 2 — VerwaltBot repo

  • Branch: review-full-verwaltbot (cherry-picked 201 commits from initial commit 597544c onto branch)
  • Comparison: review-full-verwaltbot → main
  • Scope: 731 files changed, +131107 / -37 LOC
  • Findings: []
  • Duration: ~20 seconds
  • Free run consumed: 2/3 → 3/3
  • Session ID: session_013EXdL7Du2PQgMVEpDWvzQD

Root cause hypothesis

Same as #50029: silent skip on large-scope bundle (>100k LOC), free run consumed despite no review delivered. No error message, no findings, no agent activity logs visible from CLI.

Request

Refund both free runs. Plus: clearer error surfacing when bundle exceeds analyzable size limit.

Environment

  • Plan: Claude Pro/Max
  • Date: 2026-05-06
  • Two separate repositories, same failure mode

Claude Model

None

Is this a regression?

I don't know

Last Working Version

_No response_

Claude Code Version

2.1.128

Platform

Anthropic API

Operating System

macOS

Terminal/Shell

Cursor

Additional Information

_No response_

View original on GitHub ↗

This issue has 4 comments on GitHub. Read the full discussion on GitHub ↗