kaiko-ai / eva
pythonmaine43e74a41.6K linessynced 3h ago
Code health
9.6out of 10Excellent
This codebase scores 9.6 out of 10 on defect risk, which we rate excellent. It also scores maintainability 9.4 and static performance risk 10.0 out of 10. The three are scored separately and never blended into one number.
- Lines of code
- 41.6K
Recent activity
All commitsCommits
No commits indexed yet.
Decisions
- Enable logging input text for HF and LiteLLM models and unify message mergingproposed · 3h ago
- Allow template context to be passed as string or sequenceproposed · 3h ago
- Adapt prompt enabling of enable_cot options to sound more naturalproposed · 3h ago
- Rename missing_response to missing_answerproposed · 3h ago
- Update raw extractor behavior and tests; add multi-character extraction supportproposed · 3h ago
- Remove ExtractAnswerFromRaw due to unstructured free-text requiring no parsingproposed · 3h ago
Needs attention
33 open- medium severity. ProposedEnable logging input text for HF and LiteLLM models and unify message mergingAuto-proposed decision awaiting review
- medium severity. ProposedAllow template context to be passed as string or sequenceAuto-proposed decision awaiting review
- medium severity. ProposedAdapt prompt enabling of enable_cot options to sound more naturalAuto-proposed decision awaiting review
- medium severity. ProposedRename missing_response to missing_answerAuto-proposed decision awaiting review
- medium severity. ProposedUpdate raw extractor behavior and tests; add multi-character extraction supportAuto-proposed decision awaiting review
Composition
Open the graphpython 82%markdown 10%yaml 7%json 1%toml 0%
Explore this codebase
- Docs570 pages across 7 modules
- ChatAsk this codebase a question and get an answer with its sources
- Files773 files with per-file docs, health and history
- ArchitectureDependency graph, layers, and 2 entry points
- Code healthPer-file scores, 540 open findings, coverage and refactoring targets
- Knowledge graphEntities, communities, and the paths between them
- Change couplingFiles that keep changing together, mined from commit history
- CommitsChange-risk ranked history with agent provenance
- ContributorsBus factor, per-file maintainers, and the human/agent split
- StatsSize class, origin, lifetime churn, rhythm and records
- CostsWhat indexing this snapshot cost, by model and by run