[Vincent Biret (baywet)]

Motivation & Context

Keep Magentic workflows paused consistently while a generated plan is awaiting review, including when a session is resumed.

Description & Review Guide

• What are the major changes? Persist pending plan-review state, ignore unrelated turns with a workflow warning, and resume coordination after the matching approval. Discard buffered ordinary messages before processing review responses, while preserving explicit revision feedback. Preserve review state when restoring older checkpoints and document the expected interaction.
• What is the impact of these changes? Initial and revised plans keep waiting for review across follow-up turns and serialized-session restores. Workflows without plan signoff and already-approved plans retain their normal progression.
• What do you want reviewers to focus on? Review-state transitions, checkpoint compatibility, and the streaming/non-streaming regression coverage.

Commands run from dotnet:

• dotnet test --project tests\Microsoft.Agents.AI.Workflows.UnitTests -f net10.0 --filter-method Microsoft.Agents.AI.Workflows.UnitTests.MagenticOrchestrationTests.PlanReview_Pending_Ignores_Unrelated_TurnsAsync --no-ansi --output Detailed --minimum-expected-tests 4: four behavioral failures before the fix; all four passed afterward.
• dotnet test --project tests\Microsoft.Agents.AI.Workflows.UnitTests -f net10.0 --filter-method Microsoft.Agents.AI.Workflows.UnitTests.MagenticOrchestrationTests.PlanReview_Response_Discards_Unrelated_Buffered_MessagesAsync --no-ansi --output Detailed --minimum-expected-tests 4: four behavioral failures before the follow-up correction; all four passed afterward.
• dotnet test --project tests\Microsoft.Agents.AI.Workflows.UnitTests -f net10.0 --no-ansi --minimum-expected-tests 884: passed, 884 tests, including the original regressions and mixed-response coverage.
• dotnet build src\Microsoft.Agents.AI.Workflows\Microsoft.Agents.AI.Workflows.csproj --no-restore --tl:off: passed across all target frameworks with zero warnings and errors.
• dotnet build src\Microsoft.Agents.AI.Workflows\Microsoft.Agents.AI.Workflows.csproj -c Release --tl:off: passed at the initial revision across all target frameworks with zero warnings and errors.
• dotnet format src\Microsoft.Agents.AI.Workflows\Microsoft.Agents.AI.Workflows.csproj --no-restore: passed.
• dotnet format tests\Microsoft.Agents.AI.Workflows.UnitTests\Microsoft.Agents.AI.Workflows.UnitTests.csproj --no-restore: passed.
• dotnet build agent-framework-dotnet.slnx --tl:off: stopped after local disk exhaustion. Solution-wide unit tests were not run.

Related Issue

N/A - maintainer-directed bugfix; no public issue.

AI Assistance

• No material AI assistance was used.
• This is an AI-assisted contribution. I reviewed, understood, and verified all submitted content and accept responsibility for it.

AI assistance details: Implementation, regression tests, documentation, validation, and this PR description.

Contribution Checklist

• The code builds clean without any errors or warnings
• All unit tests pass, and I have added new tests where possible
• The PR follows the Contribution Guidelines
• This PR links to an agreed issue with no competing open PR, or the Related Issue section documents a trivial-change or repository-automation exception.
• This is not a breaking change. If it is a breaking change, add the breaking change label (or add "[BREAKING]" to the title prefix, before or after any language prefix) — a workflow keeps the label and title prefix in sync automatically.

[github-actions]

MAF Automated Review — Iteration 1

Result: Findings reported
Scope: full PR (1 commit(s)): 749f508b5561
Model: gpt-5.6-sol

Overview

The change adds an explicit, checkpointed pending-review flag and uses it to reject ordinary turns until the active review is answered; request-ID correlation and restore tests provide strong protection against stale responses and checkpoint regressions. However, ordinary content submitted in the same invocation as an approval is buffered before the approval handler clears the flag, allowing that pre-approval content to enter coordination on a later turn.

Reviewed the supplied pull-request change set across correctness, security/reliability, architecture, and failure behavior.
1 verified finding remained after source verification (1 medium) across 1 file. Details are attached to the affected lines below.

Affected areas: dotnet/src/Microsoft.Agents.AI.Workflows/Specialized/Magentic/MagenticOrchestrator.cs

[Copilot]

🟢 Approval recommended

No active review issues are reported for the changes preserving pending plan reviews across workflow turns and session restoration.

No open issues 🎉

_{Review effort: Balanced}

[github-code-quality]

Code coverage overview

Languages: C#

C# / code-coverage/dotnet

main 6afcecd	fix/magentic-pending... e582d4c	+/-
86%	86%	0%

View coverage of the 3 most impacted files.

File	main 6afcecd	fix/magentic-pending... e582d4c	+/-
/home/runner/wo...SessionStore.cs	100%	56%	-44%
/home/runner/wo...Orchestrator.cs	97%	97%	0%
/home/runner/wo...kflowSession.cs	87%	87%	0%

ℹ️ Calculations represent line coverage. Results may change as calculations are updated.

---

_{Updated October 28, 2026 14:47 UTC}

[Copilot]

🟢 Approval recommended

No active review issues block approval of the changes that keep Magentic workflows paused pending plan review.

No open issues 🎉

_{Review effort: Balanced}

[github-actions]

MAF Automated Review — Iteration 2

Result: No findings
Scope: 1 net-new commit(s): e582d4c0a4a4
Model: gpt-5.6-sol

Overview

The incremental change clears executor-local ordinary messages before handling a correlated plan-review response while preserving explicit revision feedback in the response payload. It reuses the stateful executor's message-processing helper and adds regression coverage across approval, revision, streaming, non-streaming, and session-restore paths. Existing request correlation and awaiting-review guards further constrain the lifecycle. No Critical, High, or Medium residual risk was established.

Reviewed the supplied incremental change set across correctness, security/reliability, architecture, and failure behavior.
No publishable findings remained after source verification for this scope.