Let an agent account for a finding GitHub cannot close by kristofferR · Pull Request #48 · kristofferR/coderabbit-queue

kristofferR · 2026-07-26T16:06:44Z

Roadmap item D — the #4 annoyance in the transcript sweep, and the one with the most damning
observed end state: "No review was ever requested for 7a16eeb." Four rounds, one PR, no review.

The deadlock

crq resolve and crq decline both act on a review thread. These findings have none:

a review-body finding (CodeRabbit's outside-diff section),
a review_skipped notice,
any collapsed body remark crq parses but GitHub never threaded.

Drain-first says a round may not start while findings for the current head are undrained. So the
round cannot start because the finding is undrained, and the finding cannot drain because nothing
can act on it. crq next returns fix forever and no review is ever requested for the head.

crq decline was already tightened to resolve by default for exactly this reason (issue #8). That
fixed the threaded case and left this one untouched.

The lever

crq dismiss <repo> <pr> <finding-id>... --reason "why this is set aside"

Dismissed findings are withheld from the action, so the round moves on. crq next reports
dismissed: N — a set-aside finding must not look like one that was never reported.

Three choices worth arguing about:

The reason is stored, not just demanded. A dismissal that discards its justification is not
auditable, so Round.Dismissed maps ID → reason rather than being a list of IDs.

Finding IDs, not node IDs. .findings[].id is a hash of the finding's own content, so unlike a
thread ID it is not globally unique — the repo and PR genuinely identify something here and are
required.

Scoped to the current head. The same content yields the same ID, so a dismissal that outlived
its head would silently swallow the finding when the next reviewer reports it again. Superseding the
round drops it, which is the rule body findings already follow: the current reviewer must report it
again.

And dismissing enqueues the PR when crq is not yet tracking the current head. That is deliberate:
no-round-for-the-head is the deadlock's own signature, so a dismissal with nowhere to live would
change nothing.

Verification

TestDismissEndsTheUnresolvableFindingDeadlock walks the whole shape: a body finding lands with no
thread, fix repeats across two calls, dismissing ends it, the finding is withheld, dismissed is
1, nothing was posted by the dismissal itself, and a push clears the record.

TestParseDismissArgs covers both --reason forms, a missing value, an empty reason, several IDs,
and an unknown flag — a typo like --resaon must fail rather than silently become a finding ID.

gofmt -l . clean, go vet ./... clean, go test ./... -count=1 green.

Local preflight returned 7 findings (all minor/trivial); all 7 are applied — target validation now
reuses repoPR so a non-positive PR fails as it does everywhere else, the post-dismiss assertion is
exact rather than decrease-only, and the docs distinguish narrowing a PR (which fixes a skipped
review's cause) from dismissing (which only records the decision).

Ref #42

Summary by CodeRabbit

New Features
- Added crq dismiss for dismissing findings without GitHub review threads, with a required reason.
- Dismissed findings are excluded from subsequent results for the current PR head.
- Reports now show the number of dismissed findings.
- Added validation to prevent dismissing threaded or unknown findings.
Documentation
- Updated command help and agent guidance with dismissal workflows and thread-less finding behavior.

crq resolve and crq decline both act on a review thread. A review-body finding, a review-skipped notice and an outside-diff remark have none, so neither command can touch them — and drain-first then blocks every future round on a finding that can never drain. The observed end state was a PR reporting that no review was ever requested for its current head, four rounds running: the round could not start because the finding was undrained, and the finding could not drain because nothing could act on it. crq dismiss <repo> <pr> <finding-id>... --reason "<why>" records that the agent judged it. Dismissed findings are withheld from the action, and crq next reports dismissed: N so nothing looks silently dropped. Three choices worth stating. The reason is stored, not just demanded — a dismissal that discards its justification is not auditable. Finding IDs are content-derived rather than GitHub node IDs, so the repo and PR are required to identify one. And the record is scoped to the round for the current head: the same content yields the same ID, so a dismissal that outlived its head would silently swallow the finding when the next reviewer reports it again. Pushing supersedes the round and clears it, which is the rule body findings already follow. Dismissing enqueues the PR when crq is not yet tracking the current head, because that is the deadlock's own signature — no round for the head at all — and a dismissal with nowhere to live would change nothing.

coderabbitai · 2026-07-26T16:06:50Z

Warning

Review limit reached

@kristofferR, you've reached your PR review limit, so we couldn't start this review.

Next review available in: 50 minutes

Enable usage-based reviews in Billing to review now. Otherwise, wait until the next included review is available.
You're only billed for reviews past your plan's rate limits ($0.25/file).

How can I continue?

After more reviews become available, a review can be triggered using the @coderabbitai review command as a PR comment. Alternatively, push new commits to this PR.

To avoid repeated limits, reduce automatic review volume by pausing incremental auto-reviews earlier, using label-based review opt-in, excluding WIP or generated PR titles, or requesting reviews manually when the PR is ready. If your team needs uninterrupted high-volume reviews, an organization admin can enable usage-based reviews.

How do review limits work?

CodeRabbit enforces per-developer PR review limits for each organization. Most developers receive the normal plan review availability.

For paid Pro and Pro+ PR reviews, CodeRabbit uses adaptive limits for sustained high-volume activity. When a developer's recent PR review activity reaches the 95th percentile or higher among CodeRabbit users, additional reviews become available more gradually as earlier reviews age out of the rolling window.

Please refer docs for additional details.

Review details

⚙️ Run configuration

Configuration used: Organization UI

Review profile: ASSERTIVE

Plan: Pro Plus

Run ID: 94f3922f-e98f-461e-a217-fadd9be4f8cf

📥 Commits

Reviewing files that changed from the base of the PR and between b0d60a0 and 3fd635e.

📒 Files selected for processing (8)

README.md
internal/crq/dismiss.go
internal/crq/dismiss_test.go
internal/crq/feedback.go
internal/crq/next_test.go
internal/crq/service.go
llms.txt
skills/coderabbit-queue/SKILL.md

📝 Walkthrough

Walkthrough

Adds crq dismiss for threadless findings, stores dismissals per PR head, excludes dismissed findings from queue output, and documents the workflow for CLI and AI-agent usage.

Changes

Threadless Finding Dismissal

Layer / File(s)	Summary
Round dismissal state `internal/state/state.go`	Rounds persist dismissal reasons and provide methods to record and query dismissed finding IDs.
Dismissal service workflow `internal/crq/dismiss.go`, `internal/crq/next_test.go`	`Service.Dismiss` validates findings, rejects threaded or unknown IDs, updates the current-head round, and syncs the result.
Dismissal reporting and queue behavior `internal/crq/feedback.go`, `internal/crq/next.go`, `internal/crq/next_test.go`	Current-head dismissed findings are removed from feedback and reported through a separate dismissal count, with coverage for head scoping and queue progress.
CLI command and usage guidance `cmd/crq/main.go`, `cmd/crq/main_test.go`, `README.md`, `llms.txt`, `skills/coderabbit-queue/SKILL.md`	The CLI parses and dispatches `crq dismiss`, while command help and agent guidance describe handling findings without review threads.

Estimated code review effort: 4 (Complex) | ~45 minutes

Sequence Diagram(s)

sequenceDiagram
  participant Agent
  participant crq CLI
  participant Service.Dismiss
  participant Round State
  participant crq next
  Agent->>crq CLI: crq dismiss repo PR finding-id --reason
  crq CLI->>Service.Dismiss: parsed dismissal request
  Service.Dismiss->>Round State: persist current-head dismissal
  Service.Dismiss-->>crq CLI: dismissal result JSON
  Agent->>crq next: request next action
  crq next->>Round State: read current-head dismissals
  Round State-->>crq next: dismissed finding IDs
  crq next-->>Agent: findings and dismissed count

Possibly related PRs

kristofferR/coderabbit-queue#10: Both changes modify the Go CLI dispatcher and help structure.

Poem

🚥 Pre-merge checks | ✅ 5

✅ Passed checks (5 passed)

Check name	Status	Explanation
Description Check	✅ Passed	Check skipped - CodeRabbit’s high-level summary is enabled.
Title check	✅ Passed	The title clearly reflects the main change: handling findings GitHub cannot close via agent action.
Docstring Coverage	✅ Passed	Docstring coverage is 100.00% which is sufficient. The required threshold is 80.00%.
Linked Issues check	✅ Passed	Check skipped because no linked issues were found for this pull request.
Out of Scope Changes check	✅ Passed	Check skipped because no linked issues were found for this pull request.

✨ Finishing Touches

🧪 Generate unit tests (beta)

Create PR with unit tests
Commit unit tests in branch feat/dismiss-unresolvable-finding

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.}

coderabbitai · 2026-07-26T16:07:30Z

✅ Action performed

Review finished.

chatgpt-codex-connector

💡 Codex Review

Here are some automated review suggestions for this pull request.

Reviewed commit: 4a2d9880fa

ℹ️ About Codex in GitHub

Your team has set up Codex to review pull requests in this repo. Reviews are triggered when you

Open a pull request for review
Mark a draft as ready
Comment "@codex review".

If Codex has suggestions, it will comment; otherwise it will react with 👍.

Codex can also answer questions or update the PR. Try commenting "@codex address that feedback".

Three review findings, all correct. Filtering happened in nextFromState, so a dismissal did nothing for crq feedback or crq loop: convergence and the exit code are computed from the full list, leaving the finding permanently actionable outside crq next. The filter now lives in Feedback, before Converged, so there is one of it. Nothing stopped dismissing a finding that HAS a thread. That would let a round converge with the thread still open, skipping the resolve/decline flow that puts the decision on the PR where the bot can answer it. A threaded finding is now refused by name, and so is an ID that is not a finding at this head at all — a stale copy-paste would otherwise record a dismissal that silences whatever later matches it. And the write is one CAS update again. Enqueueing first meant that a failure between the enqueue and the record left a fire-eligible round behind with nothing dismissed on it, which the autoreview daemon could claim and spend a review on. Dismiss now reads the findings, validates them, then creates or supersedes the round and records the decision in a single write.

coderabbitai · 2026-07-26T16:31:02Z

✅ Action performed

Review finished.

chatgpt-codex-connector

💡 Codex Review

Here are some automated review suggestions for this pull request.

Reviewed commit: b0d60a0ec2

ℹ️ About Codex in GitHub

Your team has set up Codex to review pull requests in this repo. Reviews are triggered when you

Open a pull request for review
Mark a draft as ready
Comment "@codex review".

If Codex has suggestions, it will comment; otherwise it will react with 👍.

Codex can also answer questions or update the PR. Try commenting "@codex address that feedback".

coderabbitai

Actionable comments posted: 2

🤖 Prompt for all review comments with AI agents

Verify each finding against current code. Fix only still-valid issues, skip the
rest with a brief reason, keep changes minimal, and validate.

Inline comments:
In `@internal/crq/dismiss.go`:
- Around line 75-111: The Dismiss flow performs its CAS write directly in
dismiss.go instead of through service.go. Extract the s.store.Update logic into
a service.go effects-executor method such as recordDismissal, have Dismiss
invoke that method, and preserve the existing round, validation, and output
behavior while keeping DryRun write-free.
- Around line 58-108: The store.Update mutation currently reuses stale
feedback.Head and current data across CAS retries. Move feedback retrieval and
finding-map construction into the Update callback using the loaded round’s head,
refresh clean IDs against that feedback, and use the refreshed head for
NewRound/Supersede and validation so retries cannot regress state or reject
against stale findings.

🪄 Autofix (Beta)

Fix all unresolved CodeRabbit comments on this PR:

Push a commit to this branch (recommended)
Create a new PR with the fixes

ℹ️ Review info

⚙️ Run configuration

Configuration used: Organization UI

Review profile: ASSERTIVE

Plan: Pro Plus

Run ID: 82b94e8d-e252-4104-98ce-f98ba0c342d5

📥 Commits

Reviewing files that changed from the base of the PR and between 03c7d2e and b0d60a0.

📒 Files selected for processing (10)

README.md
cmd/crq/main.go
cmd/crq/main_test.go
internal/crq/dismiss.go
internal/crq/feedback.go
internal/crq/next.go
internal/crq/next_test.go
internal/state/state.go
llms.txt
skills/coderabbit-queue/SKILL.md

📜 Review details

🧰 Additional context used

📓 Path-based instructions (2)

internal/crq/**/*.go

📄 CodeRabbit inference engine (AGENTS.md)

Files:

internal/crq/next_test.go
internal/crq/next.go
internal/crq/dismiss.go
internal/crq/feedback.go

internal/state/**/*.go

📄 CodeRabbit inference engine (AGENTS.md)

Files:

internal/state/state.go

🔇 Additional comments (9)

Six review findings, all correct, and three of them P1. A round tracking a different head is now refused rather than superseded. If the head moved after Dismiss read the findings, superseding archives the live round and points the queue back at a commit nobody is looking at — and CAS retries cannot protect against a mutation that deliberately overwrites newer state. A round may only be CREATED when this call drains the head. Dismissing one of several open findings used to queue a fire-eligible round that DecideFire cannot hold back, because it never sees findings — so a pump could spend the primary's quota on code the caller is still expected to fix. A finding carried from an older commit is refused. IDs hash the text, not the commit, so recording one against the current head would silently filter the identical finding when the current reviewer reports it. And "no thread ID" turns out not to be the question. When the GraphQL thread query fails, Feedback falls back to REST, which does not return thread IDs at all, so an inline comment with an open thread arrives looking threadless. Only sources that intrinsically cannot have one — review bodies, prompt blocks, skip notices, issue comments — are dismissible now. The CAS write moves to service.go as recordDismissal, which is where AGENTS.md says every write belongs, and it honours DryRun.

kristofferR · 2026-07-26T17:02:16Z

@coderabbitai review

coderabbitai · 2026-07-26T17:02:22Z

✅ Action performed

Review finished.

coderabbitai · 2026-07-26T17:33:31Z

✅ Action performed

Review finished.

chatgpt-codex-connector

💡 Codex Review

Here are some automated review suggestions for this pull request.

Reviewed commit: c401ac3ded

ℹ️ About Codex in GitHub

Your team has set up Codex to review pull requests in this repo. Reviews are triggered when you

Open a pull request for review
Mark a draft as ready
Comment "@codex review".

If Codex has suggestions, it will comment; otherwise it will react with 👍.

Codex can also answer questions or update the PR. Try commenting "@codex address that feedback".

Three more findings, one P1. The eligibility guard asked whether a round object existed, when what matters is whether a FIRE-ELIGIBLE one would. A queued round is exactly as dangerous as a newly created one: Pump can hand it to DecideFire, which sees no findings and cannot enforce drain-first, so dismissing one of several open findings could still buy a review of code the caller is still fixing. Repeating a dismissal failed on its own earlier success: Feedback filters the dismissed ID out, so validating against the current findings alone rejected it as unknown. The command is documented as idempotent and an interrupted agent repeating itself is the ordinary case, so an ID this round already dismissed is accepted and reported as such. And the filter now checks the SOURCE, not just the ID. Finding IDs hash the text, not where it came from, so a dismissed body finding later delivered as an inline comment through the REST fallback hashes the same — and an ID-only filter would hide a review thread that is open.

kristofferR · 2026-07-26T17:47:41Z

@codex review

kristofferR · 2026-07-26T17:52:44Z

@coderabbitai review

coderabbitai · 2026-07-26T17:52:49Z

✅ Action performed

Review finished.

chatgpt-codex-connector

💡 Codex Review

Here are some automated review suggestions for this pull request.

Reviewed commit: c1f63aa558

ℹ️ About Codex in GitHub

Your team has set up Codex to review pull requests in this repo. Reviews are triggered when you

Open a pull request for review
Mark a draft as ready
Comment "@codex review".

If Codex has suggestions, it will comment; otherwise it will react with 👍.

Codex can also answer questions or update the PR. Try commenting "@codex address that feedback".

kristofferR · 2026-07-26T18:42:19Z

@coderabbitai review

coderabbitai · 2026-07-26T18:42:25Z

✅ Action performed

Review finished.

Three findings, one P1 in the fix itself. After a push the stored round is still on the PREVIOUS head, because `crq next` returns on a current-head finding before it enqueues. Treating that as a concurrent move rejected every dismissal and left the new head in the exact drain-first deadlock this command exists to end. The two cases are told apart by when the round was enqueued: after the findings were read means somebody moved the PR forward and the decision is stale; otherwise it is the ordinary post-push round, and it is superseded. A finding another agent dismissed concurrently now counts as handled when deciding whether other work is still open, instead of making this call refuse over a finding that is already accounted for. And the docs no longer send an agent to a command that will refuse it: a `review_comment` finding lost its thread ID to the REST fallback, so it is neither resolvable nor dismissible until crq can read review threads again.

kristofferR · 2026-07-26T19:29:35Z

@coderabbitai review

coderabbitai · 2026-07-26T19:29:40Z

✅ Action performed

Review finished.

kristofferR · 2026-07-26T19:30:29Z

@codex review

chatgpt-codex-connector

💡 Codex Review

Here are some automated review suggestions for this pull request.

Reviewed commit: c5c2e86c24

ℹ️ About Codex in GitHub

Your team has set up Codex to review pull requests in this repo. Reviews are triggered when you

Open a pull request for review
Mark a draft as ready
Comment "@codex review".

If Codex has suggestions, it will comment; otherwise it will react with 👍.

Codex can also answer questions or update the PR. Try commenting "@codex address that feedback".

Three findings, one P1 in the guard I added last round. FireEligible answers "right now", which is the wrong question when deciding whether a partial dismissal would leave a round able to buy a review. A round cooling in awaiting_retry is not eligible this second and becomes eligible the moment RetryAt passes, so letting the dismissal through on that technicality just deferred the hazard to the cooldown expiring. The guard asks whether the round will ever fire again. State alone could not tell "no round for this head" from "the head moved and nothing has enqueued it yet", so a push landing between Feedback and the write could record the dismissal against the wrong commit. The head is re-read once before recording. And the README now carries the same caveat as the skill: a review_comment finding lost its thread ID to the REST fallback and still has an open thread, so it is neither resolvable nor dismissible until crq can read threads again.

kristofferR · 2026-07-26T19:53:37Z

@codex review

chatgpt-codex-connector

💡 Codex Review

Here are some automated review suggestions for this pull request.

Reviewed commit: a3da74ebde

ℹ️ About Codex in GitHub

Your team has set up Codex to review pull requests in this repo. Reviews are triggered when you

Open a pull request for review
Mark a draft as ready
Comment "@codex review".

If Codex has suggestions, it will comment; otherwise it will react with 👍.

Codex can also answer questions or update the PR. Try commenting "@codex address that feedback".

kristofferR · 2026-07-26T20:06:32Z

@coderabbitai review

coderabbitai · 2026-07-26T20:06:37Z

✅ Action performed

Review finished.

kristofferR · 2026-07-26T21:05:07Z

@coderabbitai review

coderabbitai · 2026-07-26T21:05:12Z

✅ Action performed

Review finished.

kristofferR · 2026-07-26T21:57:41Z

@coderabbitai review

coderabbitai · 2026-07-26T21:57:46Z

✅ Action performed

Review finished.

kristofferR · 2026-07-26T22:48:54Z

@codex review

kristofferR · 2026-07-26T22:51:21Z

@coderabbitai review

coderabbitai · 2026-07-26T22:51:27Z

✅ Action performed

Review finished.

chatgpt-codex-connector Bot reviewed Jul 26, 2026

View reviewed changes

Comment thread internal/crq/dismiss.go Outdated

Comment thread internal/crq/next.go Outdated

Comment thread internal/crq/dismiss.go Outdated

chatgpt-codex-connector Bot reviewed Jul 26, 2026

View reviewed changes

Comment thread internal/crq/dismiss.go Outdated

Comment thread internal/crq/dismiss.go Outdated

Comment thread internal/crq/dismiss.go Outdated

Comment thread internal/crq/dismiss.go Outdated

coderabbitai Bot reviewed Jul 26, 2026

View reviewed changes

Comment thread internal/crq/dismiss.go Outdated

Comment thread internal/crq/dismiss.go Outdated

chatgpt-codex-connector Bot reviewed Jul 26, 2026

View reviewed changes

Comment thread internal/crq/dismiss.go

Comment thread internal/crq/feedback.go Outdated

Comment thread internal/crq/service.go Outdated

chatgpt-codex-connector Bot reviewed Jul 26, 2026

View reviewed changes

Comment thread internal/crq/service.go Outdated

Comment thread internal/crq/dismiss.go

Comment thread llms.txt Outdated

kristofferR mentioned this pull request Jul 26, 2026

Delete the trigger comments nothing needs any more #56

Open

chatgpt-codex-connector Bot reviewed Jul 26, 2026

View reviewed changes

Comment thread internal/crq/service.go Outdated

Comment thread internal/crq/service.go

Comment thread README.md

chatgpt-codex-connector Bot reviewed Jul 26, 2026

View reviewed changes

Comment thread internal/crq/dismiss.go

Dismiss the same repository it validated

3fd635e

Conversation

kristofferR commented Jul 26, 2026 • edited by coderabbitai Bot Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

The deadlock

The lever

Verification

Summary by CodeRabbit

Uh oh!

coderabbitai Bot commented Jul 26, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Review limit reached

Walkthrough

Changes

Sequence Diagram(s)

Uh oh!

coderabbitai Bot commented Jul 26, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

chatgpt-codex-connector Bot left a comment

Choose a reason for hiding this comment

💡 Codex Review

Uh oh!

Uh oh!

Uh oh!

Uh oh!

coderabbitai Bot commented Jul 26, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

chatgpt-codex-connector Bot left a comment

Choose a reason for hiding this comment

💡 Codex Review

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

coderabbitai Bot left a comment

Choose a reason for hiding this comment

Uh oh!

Uh oh!

Uh oh!

kristofferR commented Jul 26, 2026

Uh oh!

coderabbitai Bot commented Jul 26, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

coderabbitai Bot commented Jul 26, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

chatgpt-codex-connector Bot left a comment

Choose a reason for hiding this comment

💡 Codex Review

Uh oh!

Uh oh!

Uh oh!

Uh oh!

kristofferR commented Jul 26, 2026

Uh oh!

kristofferR commented Jul 26, 2026

Uh oh!

coderabbitai Bot commented Jul 26, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

chatgpt-codex-connector Bot left a comment

Choose a reason for hiding this comment

💡 Codex Review

Uh oh!

Uh oh!

Uh oh!

Uh oh!

kristofferR commented Jul 26, 2026

Uh oh!

coderabbitai Bot commented Jul 26, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

kristofferR commented Jul 26, 2026

Uh oh!

coderabbitai Bot commented Jul 26, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

kristofferR commented Jul 26, 2026 •

edited by coderabbitai Bot

Loading

coderabbitai Bot commented Jul 26, 2026 •

edited

Loading

coderabbitai Bot commented Jul 26, 2026 •

edited

Loading

coderabbitai Bot commented Jul 26, 2026 •

edited

Loading

coderabbitai Bot commented Jul 26, 2026 •

edited

Loading

coderabbitai Bot commented Jul 26, 2026 •

edited

Loading

coderabbitai Bot commented Jul 26, 2026 •

edited

Loading

coderabbitai Bot commented Jul 26, 2026 •

edited

Loading

coderabbitai Bot commented Jul 26, 2026 •

edited

Loading

coderabbitai Bot commented Jul 26, 2026 •

edited

Loading

coderabbitai Bot commented Jul 26, 2026 •

edited

Loading

coderabbitai Bot commented Jul 26, 2026 •

edited

Loading

coderabbitai Bot commented Jul 26, 2026 •

edited

Loading