agentscope-ai/agentscope v1.0.1 on GitHub

What's Changed

Fix the bug in the handle_interrupt method of ReActAgent
Fix the bug in formatter that doesn't support string output in tool result block
Add examples in AgentScope repo
- Meta Planner
- Nine-players Werewolves Game
- MCP
- Multi-agent conversation
- Deep Research agent
Support model-level structured output

[Hotfix] Fix error in pypi workflow by @DavdGao in #668
Fix hyperlink by @qbc2016 in #669
Add MCP Usage Demo with Structured Output by @LiangYang666 in #672
Add sate.yml and improve the current pull request template by @DavdGao in #675
Add an example for multi-agent conversation by @DavdGao in #673
Add base_http_api_url in DashScopeChatModel's constructor by @DavdGao in #677
Support both v0 and v1 in agentscope documentation by @DavdGao in #686
[Hotfix] Fix error in sphinx documentation by @DavdGao in #687
add deep research agent as an example by @Luohh5 in #679
[hotfix] fix the typo in longterm-memory tutorial by @Osier-Yi in #688
Meta Planner by @ZiTao-Li in #676
Implement werewolf by @qbc2016 in #681
Implement model structured output, and add model test by @qbc2016 in #678
Enriched the current README.md and README_zh.md; Fix the bug in the handle_interrupt method of ReActAgent by @DavdGao in #698
[HOTFIX] Fix typo in README.md and README_zh.md by @DavdGao in #699

Full Changelog: v1.0.0...v1.0.1