Reviewers learned from past reviews
Mines a repository's review discussions for recurring feedback and turns each cluster into a reusable review instruction.
Sub-techniques
4
| ID | Name | Description |
|---|---|---|
V1037.001 | Discussion mining | Collect resolved review threads with the code they were on. |
V1037.002 | Topic clustering | Cluster comments into themes with a minimum number of supporting discussions. |
V1037.003 | Instruction distillation | Write one instruction per theme with examples from the threads. |
V1037.004 | Memory of facts and rules | Extract facts from new discussions and match them to existing rules to update or add them. |
Craft it
- Export review threads with their file paths and diffs.
- Embed and cluster comments; keep clusters with enough independent threads.
- Distill each cluster into an instruction and evaluate it on held-out PRs.
Reverse engineer it
Signs that a codebase already runs this technique, and where to read how it is configured.
- The instructions on this site are this technique applied to each source repository.
Open-source packages
5
- Awesome Reviewers baz-scm/awesome-reviewers
- BERTopic MaartenGr/BERTopic
- HDBSCAN scikit-learn-contrib/hdbscan
- Sentence Transformers huggingface/sentence-transformers
- Mem0 mem0ai/mem0