Refactor AgentScope ReAct Agent workflow example #303

pan-x-c · 2025-09-26T08:40:16Z

Description

[Please describe the background, purpose, changes made, and how to test this PR]

Checklist

Please check the following items before code is ready to be reviewed.

Code has passed all tests
Docstrings have been added/updated in Google Style
Documentation has been updated
Code is ready for review

gemini-code-assist · 2025-09-26T08:40:33Z

Summary of Changes

Hello @pan-x-c, I'm Gemini Code Assist¹! I'm currently reviewing this pull request and will post my feedback shortly. In the meantime, here's a summary to help you and other reviewers quickly get up to speed!

This pull request refactors and expands the workflow examples by integrating the AgentScope framework, specifically demonstrating the use of its ReAct agent. It provides a structured approach to defining agent behaviors, handling responses, and calculating rewards for various tasks, starting with a concrete example for mathematical problem-solving.

Highlights

New AgentScope Integration: Introduces a dedicated directory and classes for integrating AgentScope workflows into the Trinity-RFT system.
ReAct Agent Implementation: Adds the AgentScopeReActAgent class, which wraps AgentScope's ReAct agent, enabling structured responses via Pydantic models and dynamic tool calling.
ReAct Workflow Example: Implements AgentScopeReActWorkflow to demonstrate how to use the ReAct agent within the Trinity-RFT system, including task solving, reward calculation, and experience construction.
Task-Specific Templates: Introduces a templating system for different task types, with an initial implementation for the GSM8K math problem-solving task, defining system prompts, response structures, and reward functions.

Using Gemini Code Assist

The full guide for Gemini Code Assist can be found on our documentation page, here are some quick tips.

Invoking Gemini

You can request assistance from Gemini at any point by creating a comment using either /gemini <command> or @gemini-code-assist <command>. Below is a summary of the supported commands on the current page.

Feature	Command	Description
Code Review	`/gemini review`	Performs a code review for the current pull request in its current state.
Pull Request Summary	`/gemini summary`	Provides a summary of the current pull request in its current state.
Comment	@gemini-code-assist	Responds in comments when explicitly tagged, both in pull request comments and review comments.
Help	`/gemini help`	Displays a list of available commands.

Customization

To customize Gemini Code Assist for GitHub experience, repository maintainers can create a configuration file and/or provide a custom code review style guide (such as PEP-8 for Python) by creating and adding files to a .gemini/ folder in the base of the repository. Detailed instructions can be found here.

Limitations & Feedback

Gemini Code Assist may make mistakes. Please leave feedback on any instances where its feedback is incorrect or counter productive. You can react with 👍 and 👎 on @gemini-code-assist comments. If you're interested in giving your feedback about your experience with Gemini Code Assist for Github and other Google products, sign up here.

You can also get AI-powered code generation, chat, as well as code reviews directly in the IDE at no cost with the Gemini Code Assist IDE Extension.

Review the Privacy Notices, Generative AI Prohibited Use Policy, Terms of Service, and learn how to configure Gemini Code Assist in GitHub here. Gemini can make mistakes, so double check it and use code with caution. ↩

gemini-code-assist

Code Review

This pull request introduces a new workflow for AgentScope's ReAct agent. The implementation is well-structured, with clear separation of the agent wrapper, the workflow logic, and task-specific templates. I've identified a few issues, including a critical bug that would cause a runtime error during reward calculation, a potential crash if a task template is not found, and some incorrect type hints that affect code clarity and correctness. My suggestions aim to fix these issues and improve the robustness and maintainability of the new workflow.

trinity/common/workflows/agentscope/react/react_workflow.py

trinity/common/workflows/agentscope/react/templates.py

yanxi-chen

Some inline comments for readme

docs/sphinx_doc/source_zh/tutorial/example_react.md

docs/sphinx_doc/assets/agentscope_gsm8k_reward.png

yanxi-chen

Some minor comments, otherwise lgtm

examples/agentscope_react/README.md

docs/sphinx_doc/source/tutorial/example_react.md

pan-x-c · 2025-09-28T07:54:38Z

/gemini review

gemini-code-assist

Code Review

This pull request is a significant and well-executed refactoring of the AgentScope ReAct agent workflow example. The changes introduce a more modular and cleaner structure by creating a dedicated AgentScopeReActWorkflow with its own agent wrapper and task templates. This greatly improves the clarity and maintainability of the example. The documentation has also been substantially updated to reflect these changes, making it much easier for developers to understand and adapt. The addition of supporting features like get_model_path in the model classes is a logical enhancement. Overall, this is an excellent improvement. I have a few suggestions to fix a bug and improve the documentation's accuracy.

trinity/common/workflows/agentscope/react/react_workflow.py

docs/sphinx_doc/source/tutorial/example_react.md

docs/sphinx_doc/source_zh/tutorial/example_react.md

pan-x-c · 2025-09-28T08:25:45Z

/unittest-module-common

github-actions · 2025-09-28T08:32:14Z

Summary

Tests 📝	Passed ✅	Failed ❌	Skipped ⏭️	Other ❓	Flaky 🍂	Duration ⏱️
30	30	0	0	0	0	316ms

Tests

Test Name	Status	Duration
tests/common/config_test.py::TestConfig::test_all_examples_are_valid	✅	34ms
tests/common/config_test.py::TestConfig::test_config_flatten	✅	1ms
tests/common/config_test.py::TestConfig::test_continue_from_checkpoint_is_valid	✅	1ms
tests/common/config_test.py::TestConfig::test_load_default_config	✅	1ms
tests/common/experience_test.py::TestEID::test_eid_properties	✅	1ms
tests/common/experience_test.py::TestExperience::test_action_mask_and_logprobs_type	✅	1ms
tests/common/experience_test.py::TestExperience::test_assertions	✅	1ms
tests/common/experience_test.py::TestExperience::test_dpo_experience	✅	1ms
tests/common/experience_test.py::TestExperience::test_gather	✅	1ms
tests/common/experience_test.py::TestExperience::test_hf_datasets_conversion	✅	1ms
tests/common/experience_test.py::TestExperience::test_multi_turn_experience	✅	1ms
tests/common/experience_test.py::TestExperience::test_serialize_deserialize	✅	1ms
tests/common/experience_test.py::TestExperience::test_single_turn_experience	✅	1ms
tests/common/experience_test.py::TestExperience::test_to_dict	✅	1ms
tests/common/experience_test.py::TestExperienceConversion::test_batch_conversion	✅	1ms
tests/common/experience_test.py::TestExperienceConversion::test_dpo_experience_batch_conversion	✅	1ms
tests/common/experience_test.py::TestExperienceConversion::test_experience_model_experience_conversion	✅	1ms
tests/common/experience_test.py::TestExperienceConversion::test_gather_experiences_with_custom_fields	✅	1ms
tests/common/experience_test.py::TestExperienceConversion::test_multiturn_experience_batch_converstion	✅	1ms
tests/common/vllm_test.py::ModelWrapperTest_0::test_generate	✅	58ms
tests/common/vllm_test.py::ModelWrapperTest_1::test_generate	✅	37ms
tests/common/vllm_test.py::ModelWrapperTest_2::test_generate	✅	46ms
tests/common/vllm_test.py::TestModelLen_0::test_model_len	✅	21ms
tests/common/vllm_test.py::TestModelLen_1::test_model_len	✅	20ms
tests/common/vllm_test.py::TestAPIServer::test_api	✅	24ms
tests/common/vllm_test.py::TestAsyncAPIServer::test_api_async	✅	24ms
tests/common/vllm_test.py::TestTokenizer::test_action_mask	✅	1ms
tests/common/vllm_test.py::TestTokenizer::test_action_mask_with_tools	✅	1ms
tests/common/vllm_test.py::TestAPIServerToolCall_0_deepseek_r1::test_api_tool_calls	✅	22ms
tests/common/vllm_test.py::TestAPIServerToolCall_1::test_api_tool_calls	✅	20ms

Github Test Reporter by CTRF 💚

add agentscope react agent workflow example

8a3ae5f

gemini-code-assist bot reviewed Sep 26, 2025

View reviewed changes

pan-x-c added 8 commits September 26, 2025 18:02

fix comments

90a8c75

fix comments

ac9febe

fix bugs in workflow

5155b01

fix agent

ca7dc54

update react tutorial zh

e61a9c0

add config

94617a9

update gsm8k curve

50045f2

fix dapo config

3c508a7

yanxi-chen reviewed Sep 28, 2025

View reviewed changes

docs/sphinx_doc/assets/agentscope_gsm8k_reward.png Show resolved Hide resolved

pan-x-c added 5 commits September 28, 2025 14:21

fix comments

8e59f14

update react example

a6e702a

update react en doc

f1e2508

update agentscope link

dbbcd53

add readme

0eb7326

yanxi-chen reviewed Sep 28, 2025

View reviewed changes

fix comments

757da40

gemini-code-assist bot reviewed Sep 28, 2025

View reviewed changes

trinity/common/workflows/agentscope/react/react_workflow.py Outdated Show resolved Hide resolved

docs/sphinx_doc/source/tutorial/example_react.md Outdated Show resolved Hide resolved

docs/sphinx_doc/source_zh/tutorial/example_react.md Outdated Show resolved Hide resolved

pan-x-c added 2 commits September 28, 2025 16:15

update doc name

8c4b213

fix comments

f697e70

yanxi-chen approved these changes Sep 28, 2025

View reviewed changes

yanxi-chen merged commit ee71226 into modelscope:main Sep 28, 2025
2 checks passed

Refactor AgentScope ReAct Agent workflow example #303

Refactor AgentScope ReAct Agent workflow example #303

Uh oh!

Conversation

pan-x-c commented Sep 26, 2025 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Description

Checklist

Uh oh!

gemini-code-assist bot commented Sep 26, 2025

Summary of Changes

Highlights

Footnotes

Uh oh!

gemini-code-assist bot left a comment

Choose a reason for hiding this comment

Code Review

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

yanxi-chen left a comment

Choose a reason for hiding this comment

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

yanxi-chen left a comment

Choose a reason for hiding this comment

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

pan-x-c commented Sep 28, 2025

Uh oh!

gemini-code-assist bot left a comment

Choose a reason for hiding this comment

Code Review

Uh oh!

Uh oh!

Uh oh!

Uh oh!

pan-x-c commented Sep 28, 2025

Uh oh!

github-actions bot commented Sep 28, 2025

Summary

Tests

Uh oh!

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

2 participants

pan-x-c commented Sep 26, 2025 •

edited

Loading