shareAI-lab / shareAI-lab/learn-claude-code
这里s06Context Compact 的第一层micro_compact还是有些简单粗暴的，后面会压缩重要信息导致模型出现幻觉问题
Open
Nobody has claimed this yet.
- Dominant language
- Python
- Stars
- 77.2k
- Forks
- 12.4k
- Avg merge
- 2d 5h
- Merged PRs (30d)
- 6
Description
使用的问题示例是：Read every Python file in the agents/ directory one by one。
第一层压缩会把前面的非read_file类型的tool result 替换为占位符，而有个bash result是find到的待read文件列表，这个result被压缩之前还能按顺序走loop依次读取文件，但是压缩后下一个模型读取的文件就是/agents/s03_errors.py，这个文件完全不存在，无疑是模型幻觉了。
不过这里应该只是考虑到教学，简化了压缩的逻辑，感觉实际使用的时候还要用到更有效的压缩策略。
Contributor guide
First steps
- Read the whole issue, then the project's contributing guide.
- Comment on the issue to say you are picking it up — it saves two people doing the same work.
- Fork the repository and make your change on a branch.
- Open a pull request that references the issue number.
Research direction
Start by locating the s06Context Compact implementation, focusing on the first-layer micro_compact behavior described in the issue. Reproduce the example of reading every Python file in agents/ and verify that compaction preserves the discovered file list and ordering, so the model does not select a nonexistent file afterward.
Written by the indexing model from the issue text.
Assessment
- Tech stack
- python
- Domain
- ai
- Issue type
- Bug
- Difficulty
- 5/5
- Estimated time
- Over a week
- Activity status
- Quiet
- Clarity
- Needs clarification
- Newbie friendliness
- 35/100