[sophia-ramsey]

Motivation & Context

An issue reported that resuming a workflow with both Host tool results and approval decisions incorrectly placed the entire batch in a user-role message. Although #8751 fixed this behavior, existing tests do not directly assert the mixed-role messages passed from AgentExecutor to its wrapped agent. This PR adds focused regression coverage for content roles, adjacent grouping, and original request order across streaming and non-streaming resumes, including partial responses and checkpoint restoration.

Description & Review Guide

• What are the major changes? An addition of test coverage for roles.
• What is the impact of these changes? Regression testing to prevent the issue from occurring again and a clear resolution for the linked issue.
• What do you want reviewers to focus on? Whether the testing suite is complete,

Related Issue

Fixes #8700

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: Used Copilot with GPT-6.1 Sol to find the PR that solved the issue.

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.

[Copilot]

🟢 Approval recommended

The tests comprehensively cover the reported regression without production-code changes.

0 open findings

What changed in this PR

Adds regression coverage for mixed Host tool results and approval decisions during workflow resume.

Changes:

• Verifies role grouping and request order across streaming, partial, and checkpointed resumes.
• Verifies approved/rejected tools execute exactly once without replaying Host results.

File	Description
python/​packages/​core/​tests/​workflow/​test_agent_executor_tool_calls.py	Adds focused mixed-resume regression tests and a capturing test agent.

🧠 Review effort: Balanced

---

💡 Add a code-review agent skill for context-aware, tailored reviews. Learn more in the docs.

[github-code-quality]

Code Coverage Overview

Languages: Python

Python / code-coverage/python

The overall line coverage in commit 34b8461 in the sramsey/python-tool-... branch remains at 92%, unchanged from commit 726dcff in the main branch.

[github-actions]

MAF Automated Review — Iteration 1

Result: No findings
Scope: full PR (1 commit(s)): 34b8461569d5
Model: gpt-5.6-sol

Overview

This test-only PR adds focused regression coverage for mixed Host-tool and approval responses across streaming, non-streaming, partial-response, and checkpoint-restoration paths. The tests deliberately reverse response submission, verify original request order and adjacent role grouping, exercise the real function loop, and assert exactly-once execution and input immutability. No publishable Critical, High, or Medium defect was established.

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