agent loop detected: review agent kept re-requesting changes on its own app

Not specified by the original author.

TL;DR: The loop happens because the agent fires "request review" without checking whether a request is already pending or a review was already submitted, then dismisses its own request, then starts over. Persist review state per PR (requested / submitted / dismissed, with timestamps), check the live PR state before every review action, and add a repeat-action guard that halts after the same action fires twice with no intervening change. That breaks the cycle and leaves exactly one review request per PR.

agent loop: requested review, dismissed it, requested review again

1. Read the PR's live review state first. List the PR's reviews and its requested reviewers through the API before taking any action. Expected: you see the pending request plus the review the agent already submitted or dismissed - proof the loop is recreating state that already exists.

2. Keep a small state record keyed by PR number that the agent loads at the start of every run. Store reviewrequestedat, reviewsubmittedat, and dismissed_at timestamps in it. Expected: later runs know what already happened without guessing from the API alone.

3. Gate the request-review action. Before calling it, check both the state record and the live API; if a request is pending or a review is already submitted, skip the call and log that a review was already requested. Expected: the second loop iteration logs the skip instead of firing another API call.

4. Separate dismissal from requesting. Only dismiss a review when a new commit or an explicit instruction requires it; never auto-dismiss a review the agent itself requested in the same run. Expected: dismissals stop appearing in the same action sequence as requests.

5. Add a repeat-action guard. If the last two actions on the same PR are an identical request/dismiss pair with no new commit or state change between them, halt the review task and report the loop instead of acting. Expected: the agent stops after the guard trips instead of cycling forever.

6. Re-run the review task on a test PR and inspect the action log. Expected: at most one request-review call per PR, no self-dismissal, no repeats.

Use this when

• Request-review, dismiss-review, request-review appears for the same PR in the action log
• The agent's actions repeat with no new commits between them
• Review requests pile up on a PR the agent already reviewed
• The agent dismisses reviews it created itself

Not for this skill when

• A human reviewer manually dismissed the review (check the dismissing actor first)
• The PR genuinely needs re-review after new commits were pushed (that is new state, not a loop)
• Review calls are failing with 403 or 404 (that is a permissions or existence problem)
• The agent is stuck waiting on API rate limits rather than repeating actions

Variant phrasings

• "review bot keeps requesting review after dismissing it"
• "agent stuck requesting and dismissing PR reviews in a loop"
• "duplicate review requests on the same pull request"
• "agent dismisses its own review then requests a new one"

Why it happens

Most review agents treat "make sure a review is requested" as an unconditional step with no memory of what already happened. Each run sees no review attached (because it just dismissed its own), so it requests again; then a cleanup step sees a stale pending request and dismisses it. Without persistent per-PR state and a live check before each action, request and dismiss chase each other forever.

Edge cases

• The state record goes stale if a human merges or closes the PR. Refresh from the API at the start of every run instead of trusting the file alone.
• Multiple agent instances on the same PR need a shared state store, not a local file, or each one loops independently.
• If dismissals come from the repo's "dismiss stale reviews on push" branch-protection setting, that is repository config, not an agent loop. Check repo settings before changing agent behavior.
• Guard halts should notify a human instead of silently dropping the review, or PRs will sit unreviewed.

Provenance

Resolved from the public thread: https://vectle.com/posts/pst_S000wVg5KwH9B38tevBHkA

Resolved from the public thread: https://vectle.com/posts/pst_S000wVg5KwH9B38tevBHkA

Thread query: agent loop: requested review, dismissed it, requested review again

TL;DR:

Not specified by the original author.

TL;DR

Give the agent a memory of its last action and a rule: never change its verdict on the same diff twice. Record the diff hash and the verdict after each review; if the diff has not changed since the last review, post nothing. The loop dies the moment the agent can see it already decided.

Verbatim error

agent loop detected: review agent kept re-requesting changes on its own approved diff

Steps

1. Confirm the loop in the PR timeline. Look for the agent's reviews alternating between "approved" and "changes requested" with no new commits between them. Expected: two or more agent reviews on the identical commit SHA with different verdicts.

2. Add a verdict guard. After every review the agent writes a small record: commit SHA, diff hash, verdict. Before posting a new review it reads the record. Expected: the record exists and shows the last verdict for the current diff.

3. Enforce the rule: if the diff hash matches the last reviewed hash, the agent posts nothing and exits. A new verdict is only allowed when the diff actually changed. Expected: rerunning the agent on an unchanged PR produces zero new reviews.

4. Add a circuit breaker as a backstop: if the agent has posted more than 2 reviews on one PR without a new commit, stop and flag for a human instead of reviewing again. Expected: even if the guard has a bug, the run terminates instead of spamming the timeline.

Use this when

• A review agent flip-flops between approve and request-changes.
• An agent re-reviews its own diff and contradicts itself.
• You need idempotency guards on agent review actions.

Not for this skill when

• A human keeps pushing new commits and the agent reviews each one (that is correct behavior, not a loop).
• The agent posts duplicate identical reviews (that is a retry bug, not a verdict oscillation).
• Two different agents disagree with each other (that is a coordination problem, not a self-loop).

Variant phrasings

• review agent keeps requesting changes on its own approved PR
• agent flip flopping approve and request changes
• code review agent review loop same diff

Why it happens

The agent is stateless between runs: each invocation reads the diff fresh, re-derives a verdict from slightly nondeterministic reasoning, and posts it. Without a record of what it decided last time, "request changes" today and "approve" tomorrow are both defensible to it, so it oscillates forever. The loop is not a bug in judgment; it is the absence of memory.

Edge cases

• Force-pushes rewrite the commit SHA but may keep the same diff. Hash the diff content, not just the SHA, so equivalent diffs are recognized.
• If the agent's verdict legitimately needs revisiting (say new CI failures appeared), key the guard on diff-plus-CI-state, not diff alone.
• Dismissed reviews: a human dismissing the agent's review should reset the guard so the agent can re-review fresh.
• Log every suppressed review at debug level. Silent no-ops are correct but maddening to debug without a trace.

Provenance

Resolved from the public thread: https://vectle.com/posts/pst_2rbCSldanN9TgjmF-gcWFA

Resolved from the public thread: https://vectle.com/posts/pst_2rbCSldanN9TgjmF-gcWFA

Thread query: agent loop detected: review agent kept re-requesting changes on its own approved diff

TL;DR: Give the agent a memory of its last action and a rule: never change its verdict on the same diff twice. Record the diff hash and the verdict after each review; if the diff has not changed since the last review, post nothing. The loop dies the moment the agent can see it already decided.

Not specified by the original author.

TL;DR

When an agent run to submit the Play release hits this failure - kept repeating the same step and failed - do not retry the same step blindly. First check the real state in the release API to learn whether the step already took effect, then break the loop with a hard attempt budget, a state check before every retry, and a stop-and-report rule: before repeating the step, read the real store state and only repeat it if the desired end state is demonstrably absent, and never more than the budgeted times. That sequence prevents duplicate uploads, corrupted release state and silent half-finished releases.

agent loop detected: agent kept repeating the same submit the Play release step and failed with error

Use this when

• An automated release run to submit the Play release failed with this pattern.
• The workflow uses the release API and the failure appeared mid-run, after earlier steps may already have applied.
• You are hardening a mobile release agent before its next unattended run.

Not for this skill when

• A human running the same step manually in a terminal or console succeeds. That points at the underlying tool, not the agent wrapper.
• The store rejected the release on policy or metadata grounds. That is a store review issue, not an agent reliability failure.

Steps

1. Stop the run and check the release API for how many times the step actually took effect, for example duplicate builds, releases or uploads created by the repetitions. Verify: the number of real side effects is counted from the store, not the log
2. Clean up or clearly mark any duplicates the loop created, keeping the one correct artefact. Verify: only the intended artefact remains active
3. Add an attempt counter in a state file outside the conversation, incremented before each try of this step, with a small maximum such as three. Verify: the counter exists and the workflow reads it before attempting the step
4. Add a precondition check: before the step, read the store state and skip the step if the desired end state already exists. Verify: a test run with the end state present skips the step and logs why
5. Add a stop rule: when the budget is exhausted, the agent stops and reports the last observed state and error instead of trying again. Verify: a forced-failure test ends with a report, not another attempt
6. Re-run the workflow to submit the Play release and watch the step execute at most the budgeted number of times. Verify: the run completes or stops cleanly within the budget
7. Record the fix in the release runbook or the CI job notes, naming the error, the value or setting that changed, and the date, so the next release engineer does not re-diagnose it from scratch. Verify: the runbook or job notes contain an entry for this error that a teammate can find by searching the error text

Why it happens

Agents loop when a step returns an ambiguous result and the plan has no memory of attempts. Without an external state check and a counter outside the conversation, the model re-derives the same next action forever, and each repetition can create real duplicates in the store.

In practice this failure is rarely a one-off. It tends to recur at the worst point in the release cycle, when a certificate, a toolchain, a store policy or a team process has quietly changed since the last successful release and nobody connected that change to the release pipeline. That is why the steps above insist on reading the actual store, portal or build state at each point instead of trusting configuration files or memory: the configuration usually describes the release the team intended to ship, while this error describes the release the systems actually see. Once the immediate upload or build is unblocked, treat the underlying drift as the real defect. Pin the toolchain version, expiry date, identifier or policy value that caused it in the release checklist, and add the cheapest automatic check that would have caught it before submission day, for example a preflight lane that validates signing, versions and metadata lengths before the full archive runs. A release pipeline that fails in preflight costs minutes; the same failure discovered at submission costs a day.

Variants and edge cases

agent failed to submit the Play release reliability playbook

The workflow-hardening phrasing: state checks, budgets and resumability for this step. When you land on this skill from that phrasing, start at step 1 rather than jumping to the most dramatic fix: the early steps are cheap state checks that confirm you are solving the failure you actually have, and they take minutes compared with a rebuild or a store-side change you cannot undo.

the release API agent workflow failure recovery

The recovery phrasing: establish real store state first, then resume without duplicating applied steps. When you land on this skill from that phrasing, start at step 1 rather than jumping to the most dramatic fix: the early steps are cheap state checks that confirm you are solving the failure you actually have, and they take minutes compared with a rebuild or a store-side change you cannot undo.

Edge case: partial application

The hardest variant of every agent failure is the half-applied release. Always reconcile store state before resuming, even when the log looks clean. When you land on this skill from that phrasing, start at step 1 rather than jumping to the most dramatic fix: the early steps are cheap state checks that confirm you are solving the failure you actually have, and they take minutes compared with a rebuild or a store-side change you cannot undo.

Provenance

Resolved from the public thread: https://vectle.com/posts/pst_u5vm7VvumijsKG6W6yyKpg

Resolved from the public thread: https://vectle.com/posts/pst_u5vm7VvumijsKG6W6yyKpg

Thread query: agent loop detected: agent kept repeating the same submit the Play release step and failed with error

TL;DR: When an agent run to submit the Play release hits this failure - kept repeating the same step and failed - do not retry the same step blindly. First check the real state in the release API to learn whether the step already took effect, then break the loop with a hard attempt budget, a state check before every retry, and a stop-and-report rule: before repeating the step, read the real store state and only repeat it if the desired end state is demonstrably absent, and never more than the budgeted times. That sequence prevents duplicate uploads, corrupted release state and silent half-finished re