Multi-model consensus PR review Squad + mode support in team.md#200
Merged
Multi-model consensus PR review Squad + mode support in team.md#200
Conversation
Replace 5 specialized reviewers (bug-hunter, security-analyst, etc.) with 5 generic reviewers that each dispatch 5 sub-agents across different models: 2× Opus 4.6, 1× Sonnet 4.6, 1× Gemini 3 Pro, 1× Codex 5.3. Each worker fetches the PR diff, fans out to 5 models in parallel, and synthesizes findings using a consensus filter (2+ models must flag an issue). The orchestrator just assigns 1 PR per worker and produces a summary table. Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>
SquadDiscovery now reads an optional 'mode:' line from team.md to set the multi-agent mode (broadcast, sequential, orchestrator, or orchestrator-reflect). Defaults to OrchestratorReflect when not specified. PR Review Squad team.md now specifies 'mode: orchestrator' so it's correct out of the box without manual mode selection. Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>
PureWeen
force-pushed
the
multi-model-pr-review
branch
from
e0db264 to
76f65eb
Compare
PureWeen
added a commit
that referenced
this pull request
Mar 25, 2026
) ## Problem The orchestrator dispatch prompt led with **"Assign ALL workers that have relevant work"** which caused models to invent micro-tasks to fill all 5 workers even for single-item requests. For example, a single PR re-review would get split into: - Worker-1: check commits - Worker-2: do 5-model dispatch - Worker-3: grep one file - Worker-4: build and test - Worker-5: grep another file This wastes resources and makes sessions appear stuck (all 5 workers busy for 10+ minutes on what should be a single-worker job). ## Fix Flip the emphasis so the **default is ONE worker** and fan-out is the exception: - `Assign the MINIMUM number of workers needed` - `Do NOT split a single task into micro-tasks across workers` - Only fan out for genuinely independent tasks (e.g., "review PR #100 and PR #200") ## History - **Mar 10** (`a79f9f40`): Introduced aggressive fan-out with retry loop - **Mar 12** (`9c58ebf0`): Removed retry loop, added "only relevant" caveat - **This PR**: Flips the default from "assign all" to "assign minimum" --------- Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>
arisng
pushed a commit
to arisng/PolyPilot
that referenced
this pull request
Apr 4, 2026
…ureWeen#429) ## Problem The orchestrator dispatch prompt led with **"Assign ALL workers that have relevant work"** which caused models to invent micro-tasks to fill all 5 workers even for single-item requests. For example, a single PR re-review would get split into: - Worker-1: check commits - Worker-2: do 5-model dispatch - Worker-3: grep one file - Worker-4: build and test - Worker-5: grep another file This wastes resources and makes sessions appear stuck (all 5 workers busy for 10+ minutes on what should be a single-worker job). ## Fix Flip the emphasis so the **default is ONE worker** and fan-out is the exception: - `Assign the MINIMUM number of workers needed` - `Do NOT split a single task into micro-tasks across workers` - Only fan out for genuinely independent tasks (e.g., "review PR PureWeen#100 and PR PureWeen#200") ## History - **Mar 10** (`a79f9f40`): Introduced aggressive fan-out with retry loop - **Mar 12** (`9c58ebf0`): Removed retry loop, added "only relevant" caveat - **This PR**: Flips the default from "assign all" to "assign minimum" --------- Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Summary
Two changes to improve the Squad-based PR review workflow.
How to use
Review PRs #194, #193, #191, #192, #190The mode is automatically set to Orchestrator via
mode: orchestratorinteam.md— no manual mode selection needed.Example output
The orchestrator produces a summary table like:
Each worker's detailed report includes only issues flagged by 2+ models (consensus filter), with file:line references and severity ratings.
1. Restructure PR Review Squad for multi-model consensus
Replaces 5 specialized reviewers (bug-hunter, security-analyst, etc.) with 5 generic reviewers that each independently perform a full multi-model consensus review.
Each worker:
gh pr diff/gh pr viewclaude-opus-4.6(deep bug analysis + architecture review)claude-sonnet-4.6(correctness + edge cases)gemini-3-pro-preview(security focus)gpt-5.3-codex(code quality + logic errors)2. Support
mode:field in team.mdSquadDiscoverynow reads an optionalmode:line fromteam.mdto set the multi-agent mode automatically. Previously hardcoded toOrchestratorReflect.Supported values:
broadcast,sequential,orchestrator,orchestrator-reflect# My Team mode: orchestratorTests
ParseModecovering all modes + case insensitivity + default behavior