feat: complexity-gated AI review triggers after CR rounds (#362)#441
Conversation
|
Your free trial PR review limit of 100 PRs has been reached. Please upgrade your plan to continue using CodeAnt AI. |
|
@cursor review |
|
Note Reviews pausedIt looks like this branch is under active development. To avoid overwhelming you with review comments due to an influx of new commits, CodeRabbit has automatically paused this review. You can configure this behavior by changing the Use the following commands to manage reviews:
Use the checkboxes below for quick actions:
📝 WalkthroughWalkthroughAdds a config-driven complexity-trigger flow: new ChangesComplexity-triggered AI Review Flow
Sequence Diagram(s)sequenceDiagram
participant Poll as Polling Cycle
participant Gate as maybe-trigger-ai-review.sh
participant Complexity as complexity-score.sh
participant Rounds as cycle-count.sh
participant Config as pm-config
participant State as session-state.json
participant GitHub as GitHub API
Poll->>Gate: invoked when /fixpr does not fire
Gate->>Config: read thresholds & cadence
Gate->>Complexity: request SCORE for PR
Complexity->>GitHub: gh pr view (additions,deletions,changedFiles)
GitHub-->>Complexity: PR metadata
Complexity-->>Gate: SCORE
Gate->>Rounds: compute CR_ROUNDS (--cr-only)
Rounds->>GitHub: fetch reviews (filter coderabbitai[bot] if cr-only)
GitHub-->>Rounds: CR_ROUNDS
Gate->>State: read per-PR trigger state
alt Gates pass
Gate->>GitHub: post reviewer mention comments and optional /pr-review-help
GitHub-->>Gate: comment success
Gate->>State: update last_head_sha, last_cr_round, clear steps
else Gates skip
Gate-->>Poll: emit skipped status
end
Estimated code review effort🎯 4 (Complex) | ⏱️ ~45 minutes Possibly related PRs
Suggested reviewers
Poem
🚥 Pre-merge checks | ✅ 4 | ❌ 1❌ Failed checks (1 warning)
✅ Passed checks (4 passed)
✏️ Tip: You can configure your own custom pre-merge checks in the settings. ✨ Finishing Touches🧪 Generate unit tests (beta)
Comment |
![coderabbitai[bot]](https://avatars.githubusercontent.com/in/347564?s=60&v=4){kind=small}
There was a problem hiding this comment.
Actionable comments posted: 2
🤖 Prompt for all review comments with AI agents
Verify each finding against the current code and only fix it if needed.
Inline comments:
In @.claude/scripts/complexity-score.sh:
- Line 20: Remove the unused variable assignment SCRIPT_DIR in the script:
locate the line that sets SCRIPT_DIR="$(cd "$(dirname "$0")" && pwd)" and delete
that assignment since SCRIPT_DIR is never referenced elsewhere; ensure no other
code relies on SCRIPT_DIR (e.g., functions or calls that expect the script
directory) and run a quick lint or shellcheck to verify no leftover references
remain.
In @.claude/scripts/maybe-trigger-ai-review.sh:
- Around line 167-174: post_comments currently posts non-idempotent GH comments
before persisting dedupe state, which can cause duplicate posts on failure;
change the flow in the post_comments function to first call the session-state
persistence helper and abort (exit non-zero) if that helper fails, then only
post comments after persistence succeeds; additionally implement per-trigger
idempotency markers (unique keys per comment type) that are stored/checked via
the same session-state helper so the script skips posting if a marker already
exists, and apply the same persistence-first + idempotency-check pattern to the
other comment-posting block that mirrors post_comments.
🪄 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: Path: .coderabbit.yaml
Review profile: ASSERTIVE
Plan: Pro
Run ID: d11b98c1-3649-4e9c-b606-6489e78eb557
📒 Files selected for processing (9)
.claude/pm-config.md.claude/rules/cr-github-review.md.claude/scripts/README.md.claude/scripts/complexity-score.sh.claude/scripts/cycle-count.sh.claude/scripts/maybe-trigger-ai-review.sh.claude/skills/fixpr/SKILL.md.claude/skills/pr-review-help/SKILL.mdREADME.md
![cursor[bot]](https://avatars.githubusercontent.com/in/1210556?s=60&v=4){kind=small}
|
@graphite-app re-review |
|
User cursoragent@cursor.com does not have a PR Review subscription. Go to Team management and add this email to the PR Review subscription. |
|
@cursor review |
There was a problem hiding this comment.
Actionable comments posted: 1
♻️ Duplicate comments (1)
.claude/scripts/maybe-trigger-ai-review.sh (1)
270-277:⚠️ Potential issue | 🟠 Major | 🏗️ Heavy liftClose the remaining idempotency gap after successful comment posts.
If
gh pr commentsucceeds but the subsequent state write fails, retry can re-post the same trigger because the step flag may still be false. Add a durable idempotency key per step (e.g., marker/comment-id lookup) before attempting a new post.🤖 Prompt for AI Agents
Verify each finding against the current code and only fix it if needed. In @.claude/scripts/maybe-trigger-ai-review.sh around lines 270 - 277, The current flow can re-post a trigger if gh pr comment succeeds but the STATE_HELPER write fails; add a durable idempotency marker per step so we check-and-set before posting and persist the marker after posting. Change the logic in maybe-trigger-ai-review.sh to: 1) before calling gh pr comment, query the session state (.prs["${PR_KEY}"].ai_review_trigger_steps["${step_key}"].comment_id or marker) via STATE_HELPER and skip posting if present; 2) when posting, capture the created comment id (or response) from gh pr comment (or use gh api to create/comment and return id); and 3) immediately persist that comment_id/marker with STATE_HELPER (updating .prs["${PR_KEY}"].ai_review_trigger_steps to include the marker) and treat a failed state write as a retryable error path (or delete the created comment) so retries won’t re-post. Use the existing symbols INIT_STEPS, step_key, PR_KEY, gh pr comment and STATE_HELPER to locate where to add the pre-check, capture of comment id, and persistent marker write.
🤖 Prompt for all review comments with AI agents
Verify each finding against the current code and only fix it if needed.
Inline comments:
In @.claude/pm-config.md:
- Around line 1-19: Add a top-level heading to the markdown and mark the fenced
config block with a language to satisfy markdownlint (fix MD041 and MD040):
insert a heading like "## Project Management Config" at the top of the file and
change the existing fenced block that contains THRESHOLD_SCORE, FIRST_CR_ROUND,
CADENCE_ROUNDS, FILE_WEIGHT, and ENABLE_PR_REVIEW_HELP to use a language
specifier (e.g., ```ini) so the block is properly highlighted and lint-clean.
---
Duplicate comments:
In @.claude/scripts/maybe-trigger-ai-review.sh:
- Around line 270-277: The current flow can re-post a trigger if gh pr comment
succeeds but the STATE_HELPER write fails; add a durable idempotency marker per
step so we check-and-set before posting and persist the marker after posting.
Change the logic in maybe-trigger-ai-review.sh to: 1) before calling gh pr
comment, query the session state
(.prs["${PR_KEY}"].ai_review_trigger_steps["${step_key}"].comment_id or marker)
via STATE_HELPER and skip posting if present; 2) when posting, capture the
created comment id (or response) from gh pr comment (or use gh api to
create/comment and return id); and 3) immediately persist that comment_id/marker
with STATE_HELPER (updating .prs["${PR_KEY}"].ai_review_trigger_steps to include
the marker) and treat a failed state write as a retryable error path (or delete
the created comment) so retries won’t re-post. Use the existing symbols
INIT_STEPS, step_key, PR_KEY, gh pr comment and STATE_HELPER to locate where to
add the pre-check, capture of comment id, and persistent marker write.
🪄 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: Path: .coderabbit.yaml
Review profile: ASSERTIVE
Plan: Pro
Run ID: 8ad5d3a9-769f-44f3-bfa0-15203a2148a7
📒 Files selected for processing (3)
.claude/pm-config.md.claude/scripts/complexity-score.sh.claude/scripts/maybe-trigger-ai-review.sh
|
@cursor review |
![codeant-ai[bot]](https://avatars.githubusercontent.com/in/646884?s=60&v=4){kind=small}
There was a problem hiding this comment.
Actionable comments posted: 1
🤖 Prompt for all review comments with AI agents
Verify each finding against the current code and only fix it if needed.
Inline comments:
In @.claude/pm-config.md:
- Around line 23-27: Update the variable bullets to include the exact validation
rules enforced by the scripts: for FIRST_CR_ROUND (and env alias
COMPLEXITY_FIRST_CR_ROUND) state it must be an integer ≥ 3 (scripts error out
otherwise), for CADENCE_ROUNDS (COMPLEXITY_CADENCE_ROUNDS) state it must be an
integer ≥ 1, and for FILE_WEIGHT (COMPLEXITY_FILE_WEIGHT) state it must be a
positive integer (scripts reject 0 and non-positive values); keep the existing
semantic descriptions but append these hard constraints so readers know the
exact validation behavior.
🪄 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: Path: .coderabbit.yaml
Review profile: ASSERTIVE
Plan: Pro
Run ID: d7167ab0-ed55-4074-ab7d-5ec5ca405118
📒 Files selected for processing (1)
.claude/pm-config.md
|
@cursor review |
![graphite-app[bot]](https://avatars.githubusercontent.com/in/158384?s=60&v=4){kind=small}
|
CodeAnt AI is running Incremental review |
|
@cursor review |
codeant-ai
Bot
added
the
size:XL
|
@coderabbitai full review |
|
@graphite-app re-review |
|
@CodeAnt-AI review |
|
@cursor review |
✅ Actions performedFull review triggered. |
|
CodeAnt AI is running the review. |
|
Addressed in a prior commit — env var precedence fixed in b344dfd: env overrides now applied after config-file block (lines 104–111 in current HEAD). Resolving. |
|
Addressed in a prior commit — hard exit (exit 4) added at lines 227–229 in b344dfd when STATE_HELPER is not executable, matching the defensive pattern in escalate-review.sh. Resolving. |
|
@coderabbitai Fixed in |
|
@coderabbitai Fixed in |
|
Tip For best results, initiate chat on the files or code changes.
|
|
Tip For best results, initiate chat on the files or code changes.
|
codeant-ai
Bot
removed
the
size:XL
codeant-ai
Bot
added
the
size:XL
|
CodeAnt AI Incremental review completed. |
There was a problem hiding this comment.
✅ Bugbot reviewed your changes and found no new issues!
Comment @cursor review or bugbot run to trigger another review on this PR
Reviewed by Cursor Bugbot for commit 8a95393. Configure here.
Cuts: - cr-github-review.md "Never:" — drop redundant "Ask 'want me to poll?'" bullet (already covered by "Always: Enter the polling loop ... do NOT ask") - cr-github-review.md watermark — drop "(different ID sequences)" parenthetical - cr-merge-gate.md code-owner bots — trim explanatory aside about GitHub not allowing author self-approval Word count 11023 -> 10994 (hard cap 11000).
|
@cursor review |
|
CodeAnt AI is running the review. |
codeant-ai
Bot
added
size:XL
Sequence DiagramThis PR adds a poll-time helper that combines a PR complexity score and CodeRabbit-only review round count, uses session state to gate and dedupe, and when thresholds are met posts three separate AI review comments (plus optional PR review help) on the PR. sequenceDiagram
participant Poller
participant TriggerScript
participant CycleCounter
participant ComplexityScorer
participant SessionState
participant GitHub
Poller->>TriggerScript: Run maybe-trigger-ai-review for PR
TriggerScript->>CycleCounter: Get CodeRabbit review rounds
TriggerScript->>ComplexityScorer: Get PR complexity score
TriggerScript->>GitHub: Fetch current head SHA
TriggerScript->>SessionState: Load prior trigger state and dedupe
alt Gates pass for rounds and complexity
TriggerScript->>SessionState: Init per step trigger flags for this SHA
TriggerScript->>GitHub: Post separate AI reviewer comments
TriggerScript->>SessionState: Record last fired round and clear step state
else Gates fail
TriggerScript-->>Poller: Return skipped with reason and metrics
end
Generated by CodeAnt AI |
|
CodeAnt AI finished running the review. |
|
CodeAnt AI is running the review. |
codeant-ai
Bot
added
size:XL
Sequence DiagramThis PR adds a helper that, on each CodeRabbit polling cycle without a fix run, checks PR complexity and CR-only review rounds and then posts separate AI-review trigger comments when gates pass, deduped via session state. sequenceDiagram
participant Poller as CR polling workflow
participant AITrigger as maybe-trigger-ai-review
participant GitHub as GitHub API
participant SessionState as Session state store
Poller->>AITrigger: Run AI review trigger for PR
AITrigger->>GitHub: Fetch CR-only review rounds and complexity score
AITrigger->>SessionState: Load prior trigger state for this PR and head
alt Gates not satisfied or duplicate
AITrigger-->>Poller: Return skipped with reason
else Gates satisfied
AITrigger->>SessionState: Record pending trigger steps for this PR and head
AITrigger->>GitHub: Post separate AI review trigger comments
AITrigger->>SessionState: Store last fired round, head, and clear steps
end
Generated by CodeAnt AI |
|
CodeAnt AI finished running the review. |
|
CodeAnt AI is running the review. |
codeant-ai
Bot
added
size:XL
Sequence DiagramThis PR adds a helper that, on each polling cycle without a /fixpr run, uses PR complexity and CodeRabbit-only review rounds to decide when to post separate AI review trigger comments, while tracking state to avoid duplicate posts. sequenceDiagram
participant Poller as Polling workflow
participant Trigger as maybe-trigger-ai-review.sh
participant Cycle as cycle-count.sh
participant Score as complexity-score.sh
participant State as Session state
participant GitHub
Poller->>Trigger: Run maybe-trigger-ai-review.sh for PR
Trigger->>Trigger: Load complexity config from pm-config and env
Trigger->>Cycle: Get CodeRabbit-only cycle count
Cycle-->>Trigger: Return CR rounds
Trigger->>Score: Compute PR complexity score
Score-->>Trigger: Return complexity score
Trigger->>State: Read last trigger state for PR and head sha
Trigger->>State: Initialize per-step trigger state for this PR and head
Trigger->>GitHub: Post separate comments for codeant, cursor, graphite and optional pr-review-help
Trigger->>State: Record completed trigger metadata and clear per-step state
Generated by CodeAnt AI |
|
CodeAnt AI finished running the review. |
|
CodeAnt AI is running the review. |
codeant-ai
Bot
added
size:XL
Sequence DiagramThis PR adds a helper that, on each polling cycle without a /fixpr run, uses PR complexity and CodeRabbit-only review rounds to decide whether to post three separate AI reviewer comments, while deduplicating triggers via session state. sequenceDiagram
participant Poller
participant AITrigger
participant CycleCounter
participant ComplexityScorer
participant SessionState
participant GitHubPR
Poller->>AITrigger: maybe-trigger-ai-review for PR
AITrigger->>CycleCounter: Get CodeRabbit-only cycle count
CycleCounter-->>AITrigger: Return CR rounds
AITrigger->>ComplexityScorer: Get PR complexity score
ComplexityScorer-->>AITrigger: Return score
AITrigger->>SessionState: Load last trigger state for PR and head
alt Gates satisfied and not previously triggered
AITrigger->>SessionState: Initialize step tracking for this PR and head
AITrigger->>GitHubPR: Post separate AI review comments (codeant, cursor, graphite, optional pr-review-help)
GitHubPR-->>AITrigger: Confirm comments created
AITrigger->>SessionState: Persist last trigger round and head, clear step tracking
else Gates not satisfied or duplicate
AITrigger-->>Poller: Report skipped with reason (rounds or complexity)
end
Generated by CodeAnt AI |
| PM_CFG="" | ||
| [[ -n "$REPO_ROOT" && -f "$REPO_ROOT/.claude/pm-config.md" ]] && PM_CFG="$REPO_ROOT/.claude/pm-config.md" | ||
| if [[ -n "$PM_CFG" ]]; then | ||
| section=$(awk '/^## Complexity triggers/,/^## / { if (/^## Complexity triggers/) next; if (/^## /) exit; print }' "$PM_CFG" 2>/dev/null || true) |
There was a problem hiding this comment.
Suggestion: The section extractor uses an awk range that starts at ## Complexity triggers and also ends on any ## ... header, so it terminates on the same header line and returns an empty section. That makes FILE_WEIGHT from pm-config.md effectively unreadable, so repo config is silently ignored unless env override is set. Change the section parsing to skip the first matching header and stop only on the next header. [incorrect condition logic]
Severity Level: Major ⚠️
- ⚠️ PR complexity score ignores FILE_WEIGHT in pm-config.
- ⚠️ maybe-trigger thresholds rely on default file weight only.Steps of Reproduction ✅
1. Open `.claude/pm-config.md` and observe the `## Complexity triggers` section with an
`ini` block defining `FILE_WEIGHT`, e.g. at lines 11–21 (`FILE_WEIGHT=5`) in
`/workspace/claude-code-config/.claude/pm-config.md`.
2. Note that `.claude/scripts/complexity-score.sh` attempts to read that section at line
62 via `section=$(awk '/^## Complexity triggers/,/^## / { if (/^## Complexity triggers/)
next; if (/^## /) exit; print }' "$PM_CFG" ...)`.
3. Consider how awk range patterns work: the range `/^## Complexity triggers/,/^## /`
starts on the `## Complexity triggers` header line and ends on the first line matching
`^## `; for this file the *same* header line matches both patterns, so the range is active
only for that single record, and awk's range logic turns the range off immediately after
that record.
4. On the header line, the inner awk block executes `if (/^## Complexity triggers/)
next;`, which skips printing the header, and because the range ends on that same line, no
lines from the fenced `ini` block (including `FILE_WEIGHT=...`) are ever in-range;
`section` is therefore always empty, so `FILE_WEIGHT` from `pm-config.md` is never applied
and the script always uses the default `FILE_WEIGHT=5` unless `COMPLEXITY_FILE_WEIGHT` is
set.Fix in Cursor | Fix in VSCode Claude
(Use Cmd/Ctrl + Click for best experience)
Prompt for AI Agent 🤖
This is a comment left during a code review.
**Path:** .claude/scripts/complexity-score.sh
**Line:** 62:62
**Comment:**
*Incorrect Condition Logic: The section extractor uses an awk range that starts at `## Complexity triggers` and also ends on any `## ...` header, so it terminates on the same header line and returns an empty section. That makes `FILE_WEIGHT` from `pm-config.md` effectively unreadable, so repo config is silently ignored unless env override is set. Change the section parsing to skip the first matching header and stop only on the *next* header.
Validate the correctness of the flagged issue. If correct, How can I resolve this? If you propose a fix, implement it and please make it concise.
Once fix is implemented, also check other comments on the same PR, and ask user if the user wants to fix the rest of the comments as well. if said yes, then fetch all the comments validate the correctness and implement a minimal fix| parse_pm_kv() { | ||
| local key="$1" | ||
| local section | ||
| section=$(awk '/^## Complexity triggers/,/^## / { if (/^## Complexity triggers/) next; if (/^## /) exit; print }' "$PM_CFG" 2>/dev/null || true) |
There was a problem hiding this comment.
Suggestion: The same awk range logic for Complexity triggers ends immediately on the section header, so config keys are never actually parsed from pm-config.md. This causes threshold/cadence settings in repo config to be ignored and the script to run with defaults unless environment variables are set. Use a parser that captures lines after the header until the next header. [incorrect condition logic]
Severity Level: Major ⚠️
- ⚠️ AI review gating ignores pm-config thresholds and cadence.
- ⚠️ Dry-run JSON reports default, not configured, thresholds.Steps of Reproduction ✅
1. Inspect `.claude/pm-config.md` at lines 11–21 and see the `## Complexity triggers`
section with `THRESHOLD_SCORE`, `FIRST_CR_ROUND`, `CADENCE_ROUNDS`, `FILE_WEIGHT`, and
`ENABLE_PR_REVIEW_HELP` defined in an `ini` block.
2. In `.claude/scripts/maybe-trigger-ai-review.sh`, notice `parse_pm_kv` at lines 77–89,
which uses the same awk range expression as `complexity-score.sh`: `section=$(awk '/^##
Complexity triggers/,/^## / { if (/^## Complexity triggers/) next; if (/^## /) exit; print
}' "$PM_CFG" ...)`.
3. Because awk range `/^## Complexity triggers/,/^## /` starts and ends on the same header
line (the line matches both `^## Complexity triggers` and `^## `), the range is active
only for that line; the `if (/^## Complexity triggers/) next;` inside the block skips it,
so no lines from the fenced `ini` block (where `THRESHOLD_SCORE`, etc. are declared) are
ever printed into `section`, leaving `section` empty.
4. When running `./.claude/scripts/maybe-trigger-ai-review.sh <pr> --dry-run --json` with
environment overrides unset, `parse_pm_kv` returns no values (because `section` is empty),
so the main script (lines 91–111) leaves `THRESHOLD_SCORE`, `FIRST_CR_ROUND`, and
`CADENCE_ROUNDS` at their hard-coded defaults instead of using the values from
`.claude/pm-config.md`, meaning repository trigger settings in config are effectively
ignored.Fix in Cursor | Fix in VSCode Claude
(Use Cmd/Ctrl + Click for best experience)
Prompt for AI Agent 🤖
This is a comment left during a code review.
**Path:** .claude/scripts/maybe-trigger-ai-review.sh
**Line:** 80:80
**Comment:**
*Incorrect Condition Logic: The same awk range logic for `Complexity triggers` ends immediately on the section header, so config keys are never actually parsed from `pm-config.md`. This causes threshold/cadence settings in repo config to be ignored and the script to run with defaults unless environment variables are set. Use a parser that captures lines after the header until the next header.
Validate the correctness of the flagged issue. If correct, How can I resolve this? If you propose a fix, implement it and please make it concise.
Once fix is implemented, also check other comments on the same PR, and ask user if the user wants to fix the rest of the comments as well. if said yes, then fetch all the comments validate the correctness and implement a minimal fix
|
CodeAnt AI finished running the review. |
2 participants
User description
Closes #362
Summary
Adds a measurable PR complexity score, CodeRabbit-only round counting (
cycle-count.sh --cr-only), andmaybe-trigger-ai-review.shto post three separate PR comments (@codeant-ai review,@cursor review,@graphite-app re-review) when complexity + round + cadence gates pass — without batching mentions. Configuration lives in.claude/pm-config.mdunder Complexity triggers (env vars override file defaults when set). The polling workflow (cr-github-review.md) documents running the helper when/fixpris not required that cycle;fixpr/pr-review-helpskills note composition with post-push triggers (#370).Review follow-up (commit b344dfd)
pm-config.mdfirst, then applyCOMPLEXITY_THRESHOLD_SCORE,COMPLEXITY_FIRST_CR_ROUND,COMPLEXITY_CADENCE_ROUNDS,COMPLEXITY_FILE_WEIGHTwhen set (BugBot).session-state.sh; persist per-stepai_review_trigger_stepsto state before eachgh pr commentwith resume; final completion clears steps (CodeRabbit).SCRIPT_DIRincomplexity-score.sh(CodeRabbit SC2034).Calibration evidence (threshold 100)
Formula:
additions + deletions + 5 × changedFiles(defaultFILE_WEIGHT=5).Sample: 25 most recently merged PRs (
gh pr list --state merged --limit 25). 18/25 (72%) would exceed score 100 — within the 20–80% band.Individual scores from that sample included small PRs at 9, 18, 20, 25 (would not trigger) and larger ones 746, 530, 494 (would trigger).
Test plan
cycle-count.sh <PR> --cr-onlyruns and respects mutual exclusion with--exclude-botscomplexity-score.sh <PR> --jsonmatches documented formulamaybe-trigger-ai-review.sh <PR> --dry-run --jsonevaluates gates (smoke: open PR skipped whencr_rounds_lt_2).github/scripts/rule-lint.shpasses (total words ≤ hard cap and ratchet cap)Notes
coderabbit review --prompt-onlynot run (CLI not installed in this environment).Summary by CodeRabbit
New Features
Improvements
Documentation
CodeAnt-AI Description
Trigger AI review comments only when PRs are complex enough and have enough CodeRabbit rounds
What Changed
@codeant-ai review,@cursor review, and@graphite-app re-reviewcycle-count.shcan count only CodeRabbit reviews, so round-based triggering uses CR activity instead of all reviewers/pr-review-helpposting can be set in repo config and overridden with environment variablesImpact
✅ Fewer unnecessary AI review comments✅ More predictable review-trigger timing✅ Cleaner CodeRabbit round tracking🔄 Retrigger CodeAnt AI Review
Details
💡 Usage Guide
Checking Your Pull Request
Every time you make a pull request, our system automatically looks through it. We check for security issues, mistakes in how you're setting up your infrastructure, and common code problems. We do this to make sure your changes are solid and won't cause any trouble later.
Talking to CodeAnt AI
Got a question or need a hand with something in your pull request? You can easily get in touch with CodeAnt AI right here. Just type the following in a comment on your pull request, and replace "Your question here" with whatever you want to ask:
This lets you have a chat with CodeAnt AI about your pull request, making it easier to understand and improve your code.
Example
Preserve Org Learnings with CodeAnt
You can record team preferences so CodeAnt AI applies them in future reviews. Reply directly to the specific CodeAnt AI suggestion (in the same thread) and replace "Your feedback here" with your input:
This helps CodeAnt AI learn and adapt to your team's coding style and standards.
Example
Retrigger review
Ask CodeAnt AI to review the PR again, by typing:
Check Your Repository Health
To analyze the health of your code repository, visit our dashboard at https://app.codeant.ai. This tool helps you identify potential issues and areas for improvement in your codebase, ensuring your repository maintains high standards of code health.