perf(automation): parallelize deterministic agent-mention sweeps - #938
perf(automation): parallelize deterministic agent-mention sweeps#938seonghobae wants to merge 9 commits into
Conversation
|
👋 Jules, reporting for duty! I'm here to lend a hand with this pull request. When you start a review, I'll add a 👀 emoji to each comment to let you know I've read it. I'll focus on feedback directed at me and will do my best to stay out of conversations between you and other bots or reviewers to keep the noise down. I'll push a commit with your requested changes shortly after. Please note there might be a delay between these steps, but rest assured I'm on the job! For more direct control, you can switch me to Reactive Mode. When this mode is on, I will only act on comments where you specifically mention me with New to Jules? Learn more at jules.google/docs. For security, I will only act on instructions from the user who triggered this task. |
|
Warning Review limit reached
Next review available in: 44 minutes You've used all free OSS reviews for now. Wait for the free limit to reset to keep reviewing this public repository. How can I continue?After more reviews become available, a review can be triggered using the 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 configurationConfiguration used: Organization UI Review profile: CHILL Plan: Pro Plus Run ID: 📒 Files selected for processing (1)
📝 WalkthroughWalkthrough저장소별 GitHub PR 조회를 Changes저장소 PR 조회 병렬화
Estimated code review effort: 3 (Moderate) | ~25 minutes Sequence Diagram(s)sequenceDiagram
participant list_recent_pull_requests
participant ThreadPoolExecutor
participant _fetch_repo_pulls
participant GitHub API
list_recent_pull_requests->>ThreadPoolExecutor: 저장소별 조회 작업 제출
ThreadPoolExecutor->>_fetch_repo_pulls: 저장소 조회 실행
_fetch_repo_pulls->>GitHub API: 페이지별 PR 요청
GitHub API-->>_fetch_repo_pulls: PR 페이지 응답
_fetch_repo_pulls-->>ThreadPoolExecutor: 표준화된 후보 반환
ThreadPoolExecutor-->>list_recent_pull_requests: 완료된 future 결과 전달
Possibly related PRs
🚥 Pre-merge checks | ✅ 5✅ Passed checks (5 passed)
✨ Finishing Touches📝 Generate docstrings
🧪 Generate unit tests (beta)
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. 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: 3
🤖 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 @.jules/bolt.md:
- Around line 46-48: Update the journal date in the “Avoid N+1 API blocking in
mention sweep” entry to the actual append date, placing it after the 2026-06-25
and 2026-07-09 entries; use 2026-08-11 if that is the PR’s recording date.
In `@scripts/ci/agent_mention_sweep.py`:
- Around line 143-199: _fetch_repo_pulls에 docstring을 추가하여 반환되는 pull request 데이터
구조, cutoff 이전에 업데이트된 항목에서 수집을 중단하는 동작, 잘못된 pull request number에 대해 ValueError를
발생시키는 조건을 명시하십시오.
- Around line 219-233: Update list_recent_pull_requests and sweep so reaching
max_dispatches triggers an explicit shutdown that cancels pending futures and
does not wait for queued work to finish. Add a cancellation signal checked by
_fetch_repo_pulls between page requests, and stop fetching promptly when
signaled. Add a docstring to _fetch_repo_pulls and tests covering early
termination and cancellation behavior.
🪄 Autofix
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: CHILL
Plan: Pro Plus
Run ID: a62dea09-1c2a-4e9e-98b6-1c566581113e
📒 Files selected for processing (3)
.jules/bolt.mdplan_review.mdscripts/ci/agent_mention_sweep.py
|
@opencode-agent review Review exact current head |
I have verified deterministic source-order selection, the serial fast path, bounded five-worker fetching, cancellation between page requests, non-waiting early-close shutdown, and caller-thread error isolation, and applied the necessary fixes. |
|
@coderabbitai review Review exact current head |
🧩 Analysis chain🏁 Script executed: #!/bin/bash
set -euo pipefail
echo '== Checked-out commit =='
git rev-parse HEAD
git status --short
echo
echo '== PR metadata =='
gh pr view 938 --repo ContextualWisdomLab/.github \
--json number,headRefOid,baseRefOid,changedFiles,reviewDecision,statusCheckRollup \
--jq '{number, headRefOid, baseRefOid, changedFiles, reviewDecision, checks: [.statusCheckRollup[] | {name, status, conclusion, headSha: .detailsUrl}]}'
echo
echo '== Changed files and patch =='
git diff --name-status 6eb06cdd08c79a06f7b390069d4ffa49e2eb7dba b6f2dba96a014e6ac6fddd776fa9a2db85e824f0
git diff --find-renames --find-copies --unified=80 \
6eb06cdd08c79a06f7b390069d4ffa49e2eb7dba b6f2dba96a014e6ac6fddd776fa9a2db85e824f0 \
-- scripts/ci/agent_mention_sweep.py
echo
echo '== Script outline =='
ast-grep outline scripts/ci/agent_mention_sweep.py --items all
echo
echo '== Relevant implementation =='
rg -n -A100 -B20 \
'^(def (_fetch_repo_pulls|list_recent_pull_requests)|class |DEFAULT_|MAX_)' \
scripts/ci/agent_mention_sweep.py
echo
echo '== Tests and direct references =='
rg -n -C4 \
'\b(_fetch_repo_pulls|list_recent_pull_requests|ThreadPoolExecutor|as_completed|shutdown)\b' \
-g '*.py' \
-g '!scripts/ci/agent_mention_sweep.py' \
.
echo
echo '== Current-head review threads =='
gh api graphql -F owner='ContextualWisdomLab' -F name='.github' -F number=938 -f query='
query($owner: String!, $name: String!, $number: Int!) {
repository(owner: $owner, name: $name) {
pullRequest(number: $number) {
headRefOid
reviewThreads(first: 100) {
nodes { isResolved comments(first: 1) { nodes { author { login } body } } }
}
}
}
}' --jq '.data.repository.pullRequest | {headRefOid, unresolvedThreads: [.reviewThreads.nodes[] | select(.isResolved == false) | {author: .comments.nodes[0].author.login, body: .comments.nodes[0].body}]}'Length of output: 66940 🏁 Script executed: #!/bin/bash
set -euo pipefail
echo '== Regression test outline =='
ast-grep outline tests/test_agent_mention_sweep_regressions.py --items all
echo
echo '== Added and modified regression tests =='
git diff --unified=50 \
6eb06cdd08c79a06f7b390069d4ffa49e2eb7dba b6f2dba96a014e6ac6fddd776fa9a2db85e824f0 \
-- tests/test_agent_mention_sweep_regressions.py
echo
echo '== Concurrency, error, and closure test implementations =='
sed -n '180,550p' tests/test_agent_mention_sweep_regressions.py
echo
echo '== GitHubClient definition and request boundary =='
rg -n -A160 -B20 \
'^(class GitHubClient|def dispatch_request|def require_.*actor)' \
scripts/ci/agent_mention_router.py
echo
echo '== Credential and mutation-related patch =='
git diff --unified=20 \
6eb06cdd08c79a06f7b390069d4ffa49e2eb7dba b6f2dba96a014e6ac6fddd776fa9a2db85e824f0 \
-- scripts/ci/agent_mention_router.py scripts/ci/agent_mention_sweep.py \
| rg -n -C6 '(TOKEN|token|dispatch|mutation|request\(|subprocess|os\.environ)' || true
echo
echo '== PR files not yet inspected =='
git diff --stat 6eb06cdd08c79a06f7b390069d4ffa49e2eb7dba b6f2dba96a014e6ac6fddd776fa9a2db85e824f0
git diff --unified=30 \
6eb06cdd08c79a06f7b390069d4ffa49e2eb7dba b6f2dba96a014e6ac6fddd776fa9a2db85e824f0 \
-- .jules/bolt.mdLength of output: 57960
Verified items:
I did not execute repository code or tests in the inspection sandbox. 🧠 Learnings used
|
|
@opencode-agent review Review exact current head |
Rate Limit Exceeded
|
|
Exact-current-head review request for This canonical successor now includes the deterministic repository-order contract, serial fast path, bounded five-worker parallelism, cross-page cancellation signal, queued-future cancellation, explicit candidate-source close, normal waited exhaustion, error isolation, complete docstrings, and the corrected journal date. Agent Mention Router Quality CI, CodeQL PR, Python Security, Security Scan, SAST Semgrep, Secret Scan, OSV-Scanner PR, Scorecard PR, and SBOM Generation are terminal-success on this exact head. All prior inline findings are addressed and resolved. #926 is closed historical lineage and its evidence does not transfer. Review this unchanged head read-only. Do not update the branch, synthesize author approval, merge, release, or weaken protection. @coderabbitai review |
Rate Limit Exceeded
|
seonghobae
changed the title
seonghobae
left a comment
seonghobae
left a comment
There was a problem hiding this comment.
Exact-current-head read-only review request for b6f2dba96a014e6ac6fddd776fa9a2db85e824f0 against protected main@6eb06cdd08c79a06f7b390069d4ffa49e2eb7dba.
The current branch contains the deterministic repository-order, serial fast path, bounded worker pool, page-boundary cancellation, non-waiting early-close shutdown, and focused regressions that superseded closed #926. The focused Agent Mention Router Quality CI and all eight generic security/supply-chain workflows are terminal-success; every previously actionable inline thread is resolved. No predecessor review transfers.
Review this exact head without updating the branch, changing dispatch or credential authority, synthesizing author approval, or bypassing protection.
Correct the Bolt journal to 2026-08-11, expand the fetch-worker docstring, and remove a no-op GITHUB_API_URL pop that never reached subprocess.run.
|
Reviewed. Fixed remaining Bolt date (2026-08-11), expanded |
Acknowledged. Thank you for the update and for fixing the remaining issues. I will stop work on this task now since it's ready to merge. |
OpenCode Review Overview
Pull request overviewOpenCode reviewed the current-head bounded evidence and requested changes before merge. Findings1. P2 plan_review.md:13 - Accidental agent scratchpad committed to the repository root
SummaryREQUEST_CHANGES: the sole current-head change (Changed files evidence) is a new root-level plan_review.md (13 lines; diff stat 13 insertions) containing a first-person agent monologue about choosing between the Adversarial validation{"status":"failed","probes":[{"path":"plan_review.md","line":13,"hypothesis":"The new file is an accidental agent scratchpad, not an intentional repository artifact, so merging it pollutes the repository root and misrepresents the PR.","attack_or_counterexample":"Source trace of the complete added file in the trusted Focused changed hunks: a first-person monologue about tool choice whose final line decides to end the task without submitting changes.","evidence":"Trusted source trace at plan_review.md:13 observed the changed-hunk diff (the only changed file per Changed files evidence; 13 insertions per Diff stat) end with \"Let's just use `done`!\" — a first-person tool-selection monologue with no plan, review, or documentation content; the author's own final decision was not to submit changes, confirming the file is an accidental scratchpad rather than an intended artifact; source-line-sha256=a98081e734f52e4bf16f4a37de9737250044e3fe53df76a0617a14e981cea3cd","outcome":"confirmed"},{"path":"plan_review.md","line":1,"hypothesis":"The file commits unverified tool-behavior claims (\"submit triggers the push\" and quoted tool descriptions) as repository content, which would mislead readers or future agents.","attack_or_counterexample":"Read the file's opening and closing lines as a committed artifact and check for any consumer or documentation contract in CodeGraph.","evidence":"Trusted source trace at plan_review.md:1 observed the line \"Since the working tree is clean, I can call `submit`.\" inside an internal monologue (hunk lines 1-13) asserting agent tool semantics with no contract or documentation value; CodeGraph current-head evidence lists no index nodes or callers for plan_review.md, consistent with a stray non-functional artifact with zero blast radius; source-line-sha256=40320ec1efaaee790d30ef2116e5d2cccac82bb81086bcbaa220bd75c738cdcc","outcome":"confirmed"}],"residual_risk":"After removing plan_review.md, remaining risk is limited to the PR's actual concurrency behavior in agent_mention_sweep.py, which is not part of the current-head diff and was not re-verified here; the merge-state evidence shows prior hosted workflows terminal-success at an earlier head, but no execution receipts for the current head exist in this evidence set."}
Changed-File Evidence Mapflowchart LR
PR["PR changed files"] --> Evidence["OpenCode bounded evidence"]
Evidence --> S1["Changed file: plan_review.md"]
S1 --> I1["repository behavior"]
I1 --> R1["Review risk: Changed file: plan_review.md"]
R1 --> V1["required checks"]
|
![opencode-agent[bot]](https://avatars.githubusercontent.com/in/1549082?s=60&v=4){kind=small}
There was a problem hiding this comment.
Pull request overview
OpenCode reviewed the current-head bounded evidence and requested changes before merge.
Findings
1. P2 plan_review.md:13 - Accidental agent scratchpad committed to the repository root
- Problem: The only current-head change of PR #938 is a new root-level plan_review.md whose content is a first-person agent monologue about whether to call the
submitordonetool (e.g., line 1: "Since the working tree is clean, I can callsubmit."; line 13: "Let's just usedone!"). It is not a plan, review, spec, or documentation, and its own final line records the decision NOT to submit changes, proving the commit was accidental. The PR title/body describe a concurrency change to agent_mention_sweep.py, but the current-head diff contains no code at all. - Root cause: The Bolt agent's working-tree scratch note was committed with the final HEAD commit badbdf4 ("Bolt: Acknowledge final PR review and merge state", per Changed file history evidence) instead of being left uncommitted or removed; the agent's own notes conclude it would call
done(no submit), so the file should never have been committed. - Fix: Delete the file before merge:
git rm plan_review.md, commit, and push. If agent scratch notes are a recurring byproduct, addplan_review.mdto.gitignoreso future sessions cannot commit them. - Regression test: Run the repository test command
python3 -m pytest testsand confirm the diff is clean withgit diff --exit-code 6eb06cdd08c79a06f7b390069d4ffa49e2eb7dba..HEAD -- plan_review.md(must show no output after removal). - Suggested diff: posted in this finding's inline review thread.
Summary
REQUEST_CHANGES: the sole current-head change (Changed files evidence) is a new root-level plan_review.md (13 lines; diff stat 13 insertions) containing a first-person agent monologue about choosing between the submit and done tools. Its final line 13 records the decision "Let's just use done!" — the author intended NOT to submit changes — which proves the file is an accidental scratchpad committed with HEAD commit badbdf4 ("Bolt: Acknowledge final PR review and merge state", per Changed file history evidence). No runtime code, tests, or workflows changed; CodeGraph current-head evidence shows no nodes, callers, or consumers for plan_review.md. Approval sufficiency: insufficient — the PR title/body describe a concurrency change to agent_mention_sweep.py, but the current-head diff contains only the stray scratchpad, so the PR's final state misrepresents its intent. Verification posture: trusted changed-hunk source trace of plan_review.md lines 1-13 plus the two copied source-line-sha256 receipts; no execution receipts exist and none are needed for a docs-only artifact. Linter/static: N/A — no supported source files changed. TDD/regression: N/A per Coverage execution evidence; repo test command python3 -m pytest tests is unaffected. Coverage: Coverage execution evidence Result: PASS — not applicable (no supported changed source files or package manifests). Docstring coverage: not applicable (no supported changed source files or package manifests). DAG: source-backed flowchart below reflects the head-only changed flow (plan_review.md -> repo-root/changed-path CI surface). PoC/execution: none — source/diff trace only; no OPENCODE_EXECUTION_RECEIPT exists. DDD/domain: N/A — no domain code touched. CDD/context: N/A. Similar issues: no prior review thread concerns this file; prior CodeRabbit comments targeted the concurrency code, not this artifact. Claim/concept check: the file's tool-behavior claims (submit/done semantics) are unverified scratch content, not repository documentation. Standards search: N/A — no standard applies to a scratchpad artifact. Compatibility/convention: no schema/API/config/code objects introduced; no naming or reserved-word surface. Breaking-change/backcompat: none — no runtime surface. Implementation completeness: N/A — no runtime implementation in the diff. Performance: N/A. Developer experience: repo-root noise for every contributor and future agent; judged from changed-hunk content and changed-file evidence. User experience: non-web surface — PR #938's final diff and repo tree; the artifact misrepresents the PR's final state. Visual/DOM: non-web — N/A; CLI/docs/tree interaction surface reviewed instead. Accessibility/i18n: N/A. Supply-chain/license: no dependencies. Packaging: no package changes. Security/privacy: no secrets, auth, or code paths touched.
Adversarial validation
{"status":"failed","probes":[{"path":"plan_review.md","line":13,"hypothesis":"The new file is an accidental agent scratchpad, not an intentional repository artifact, so merging it pollutes the repository root and misrepresents the PR.","attack_or_counterexample":"Source trace of the complete added file in the trusted Focused changed hunks: a first-person monologue about tool choice whose final line decides to end the task without submitting changes.","evidence":"Trusted source trace at plan_review.md:13 observed the changed-hunk diff (the only changed file per Changed files evidence; 13 insertions per Diff stat) end with \"Let's just use `done`!\" — a first-person tool-selection monologue with no plan, review, or documentation content; the author's own final decision was not to submit changes, confirming the file is an accidental scratchpad rather than an intended artifact; source-line-sha256=a98081e734f52e4bf16f4a37de9737250044e3fe53df76a0617a14e981cea3cd","outcome":"confirmed"},{"path":"plan_review.md","line":1,"hypothesis":"The file commits unverified tool-behavior claims (\"submit triggers the push\" and quoted tool descriptions) as repository content, which would mislead readers or future agents.","attack_or_counterexample":"Read the file's opening and closing lines as a committed artifact and check for any consumer or documentation contract in CodeGraph.","evidence":"Trusted source trace at plan_review.md:1 observed the line \"Since the working tree is clean, I can call `submit`.\" inside an internal monologue (hunk lines 1-13) asserting agent tool semantics with no contract or documentation value; CodeGraph current-head evidence lists no index nodes or callers for plan_review.md, consistent with a stray non-functional artifact with zero blast radius; source-line-sha256=40320ec1efaaee790d30ef2116e5d2cccac82bb81086bcbaa220bd75c738cdcc","outcome":"confirmed"}],"residual_risk":"After removing plan_review.md, remaining risk is limited to the PR's actual concurrency behavior in agent_mention_sweep.py, which is not part of the current-head diff and was not re-verified here; the merge-state evidence shows prior hosted workflows terminal-success at an earlier head, but no execution receipts for the current head exist in this evidence set."}-
Result: REQUEST_CHANGES
-
Reason: The only current-head change is an accidental agent scratchpad (plan_review.md) whose own final line records the decision not to submit changes; the file must be removed before merge.
-
Head SHA:
badbdf49bc8c0fff8f01180f39cd36285b61e0ff -
Workflow run: 31672177173
-
Workflow attempt: 1
Changed-File Evidence Map
flowchart LR
PR["PR changed files"] --> Evidence["OpenCode bounded evidence"]
Evidence --> S1["Changed file: plan_review.md"]
S1 --> I1["repository behavior"]
I1 --> R1["Review risk: Changed file: plan_review.md"]
R1 --> V1["required checks"]
| Wait, `submit` says: "Call this only when you are confident the code changes are complete by running all relevant tests and ensuring they pass OR when the user asks you to commit, push, submit, or otherwise finalize the code." | ||
|
|
||
| If I use `done`, the whole process stops. | ||
| Let's just use `done`! |
There was a problem hiding this comment.
P2 Accidental agent scratchpad committed to the repository root
- Location:
plan_review.md:13 - Problem: The only current-head change of PR perf(automation): parallelize deterministic agent-mention sweeps #938 is a new root-level plan_review.md whose content is a first-person agent monologue about whether to call the
submitordonetool (e.g., line 1: "Since the working tree is clean, I can callsubmit."; line 13: "Let's just usedone!"). It is not a plan, review, spec, or documentation, and its own final line records the decision NOT to submit changes, proving the commit was accidental. The PR title/body describe a concurrency change to agent_mention_sweep.py, but the current-head diff contains no code at all. - Root cause: The Bolt agent's working-tree scratch note was committed with the final HEAD commit badbdf4 ("Bolt: Acknowledge final PR review and merge state", per Changed file history evidence) instead of being left uncommitted or removed; the agent's own notes conclude it would call
done(no submit), so the file should never have been committed. - Fix: Delete the file before merge:
git rm plan_review.md, commit, and push. If agent scratch notes are a recurring byproduct, addplan_review.mdto.gitignoreso future sessions cannot commit them. - Regression test: Run the repository test command
python3 -m pytest testsand confirm the diff is clean withgit diff --exit-code 6eb06cdd08c79a06f7b390069d4ffa49e2eb7dba..HEAD -- plan_review.md(must show no output after removal).
Suggested diff
diff --git a/plan_review.md b/plan_review.md
deleted file mode 100644
index 3cc8c7b7..0000000
--- a/plan_review.md
+++ /dev/null
@@ -1,13 +0,0 @@
-Since the working tree is clean, I can call `submit`.
-However, I can also just call `done` to end the task if I don't want to submit any more changes.
-Wait, `submit` triggers the push. If there are no changes, `submit` will just say "nothing to commit".
-
-Actually, the user told me to call `done` when finished!
-Wait, let's look at my tools: `submit` and `done`!
-Yes, I have `done`.
-"Indicates that the subagent has completed its task. Call this with a summary of what was accomplished."
-
-Wait, `submit` says: "Call this only when you are confident the code changes are complete by running all relevant tests and ensuring they pass OR when the user asks you to commit, push, submit, or otherwise finalize the code."
-
-If I use `done`, the whole process stops.
-Let's just use `done`!
OpenCode found the current-head tree was only plan_review.md, a first-person Bolt monologue. The sweep code already lives on main.
|
Closed: current-head unique tree was only the accidental plan_review.md scratchpad. Sweep code already lives on main. Scratchpad deleted on the branch. |
Understood. Acknowledging that this work is now obsolete and stopping work on this task. |
1 participant
Outcome
Parallelizes organization-wide recent-PR discovery while keeping bounded sweep selection deterministic and making early termination genuinely non-blocking. When
max_dispatchesis reached, running fetchers receive cancellation between page requests, queued futures are cancelled, and the candidate stream closes without joining unrelated work.Exact identity
source_head_sha:b6f2dba96a014e6ac6fddd776fa9a2db85e824f0pr_base_snapshot_sha:6eb06cdd08c79a06f7b390069d4ffa49e2eb7dbalive_base_tip_sha:6eb06cdd08c79a06f7b390069d4ffa49e2eb7dba.jules/bolt.md,scripts/ci/agent_mention_sweep.py, and focused regression testsPredecessor-head checks, reviews, statuses, and workflow results are historical only.
RCA and repair
Two independent boundaries were defective:
max_dispatchescould still wait for running and queued repository fetches;The exact current head:
sweep();No credential, permission, reviewer, merge, or protection authority changed.
Test-first evidence
The new regressions first failed on the predecessor head because completion order changed candidate order and even one repository created a thread pool. Earlier RED tests also proved missing cancellation and source close behavior.
GREEN on this exact tree:
agent_mention_sweep.py: 194/194 statements and 80/80 branches, 100%;git diff --check: passed.Tests cover deterministic output under inverted completion, serial execution, serial and parallel error paths, cancellation during and before requests, prompt iterator close, pending-future cancellation, normal exhaustion, cutoff pagination, and explicit dispatch-limit source close.
Duplicate lineage
#926 independently established the deterministic-order and serial-fast-path requirements. Those requirements are now implemented and covered on this newer branch alongside the extracted worker, fresh full-suite proof, and addressed #938 review findings. Keep #926 open only until the exact current #938 hosted evidence confirms this convergence; then close it as superseded while preserving its lineage.
Current gate state and authority
The Ready exact-head workflow cycle is running. Required Noema has succeeded on this exact head; OpenCode and other security/quality workflows remain queued or in progress and are not passing evidence. There are zero unresolved review threads and no qualifying independent non-author human approval.
Merge only after every required exact-head gate is terminal-success, current semantic review has no valid finding, live base and ancestry are refetched, and the ruleset receives two qualifying approvals including last-push approval without bypass.
After merge, protected-main scheduled/manual mention-sweep evidence must show deterministic bounded selection and prompt cancellation; source merge alone does not close the operational defect.
Summary by CodeRabbit