fix: general_text_memory fails due to unexpected type returned by LLM #138
Description
Pre-submission checklist | 提交前检查
- I have searched existing issues and this hasn't been mentioned before | 我已搜索现有问题，确认此问题尚未被提及
- I have read the project documentation and confirmed this issue doesn't already exist | 我已阅读项目文档并确认此问题尚未存在
- This issue is specific to MemOS and not a general software issue | 该问题是针对 MemOS 的，而不是一般软件问题
Bug Description | 问题描述
Description:
When using general_text_memory to extract structured memories, the prompt does not strictly constrain the type field returned by the LLM. As a result, the model may return values such as "goal" or "plan" instead of the expected literal values, causing pydantic validation to fail with a Literal mismatch. This leads to the entire memory item being discarded.
How to Reproduce | 如何重现
How to Reproduce:
1. Use a general prompt (e.g., SIMPLE_STRUCT_MEM_READER_PROMPT) to extract memory.
2. Do not explicitly restrict the output type in the prompt.
3. LLM may return non-allowed types like "goal" or "plan".
4. Pydantic throws a validation error like:
Environment | 环境信息
memos v0.2.1
Additional Context | 其他信息
No response
Willingness to Implement | 实现意愿
- I'm willing to implement this myself | 我愿意自己解决
- I would like someone else to implement this | 我希望其他人来解决