🤖 perf: short-circuit no-issue deep reviews#3561
Conversation
|
@codex review |
There was a problem hiding this comment.
💡 Codex Review
Here are some automated review suggestions for this pull request.
Reviewed commit: d4876bb9b8
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".
|
@codex review Please take another look after the overstated-verdict fix. |
|
Codex Review: Didn't find any major issues. Bravo. Reviewed commit: ℹ️ About Codex in GitHubYour team has set up Codex to review pull requests in this repo. Reviews are triggered when you
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". |
Summary
Short-circuits deep-review workflow runs when earlier phases prove there are no actionable findings, avoiding unnecessary triage, verification, and final-synthesis agents.
Background
A no-issue review currently still spends extra agent turns on dedupe/triage and final reporting. This keeps the structured result shape but exits as soon as no later agent can add value.
Implementation
Validation
make static-checkbun scripts/gen_builtin_workflows.ts checkbun test src/node/services/workflows/builtInWorkflowDefinitions.test.ts -t "built-in deep-review-workflow"make typecheckmake lintmake fmt-checkgit diff --checkRisks
Low-to-medium workflow orchestration risk: this changes which phases are recorded in no-issue paths, but preserves the existing structured output shape and adds regression coverage for the skipped-agent behavior.
Generated with
mux• Model:openai:gpt-5.5• Thinking:xhigh• Cost:$7.63