juliusbrussee / caveman
Code health
This codebase scores 6.4 out of 10 for code health, which we rate fair. It also scores maintainability 8.1 and static performance 9.4 out of 10. The three are scored separately and never blended into one number. Risk is concentrated, as it usually is: 9 of 143 files are git hotspots, and they average 3.8 — which is where the fixes pay off most.
Do next
2 things worth doing this quarter.
Worth planning
Break up installOpencode (CCN 48, 155 lines)
bin/install.js:783
installOpencode nests its control flow deeply, and part of it is duplicated, so splitting it also removes a copy; changed 20 times in 90 days.
- gain
- up to +0.5 health on this file(inferred)
- finding
- Nested complexity
- size
- CCN 48, 155 lines, nests 7 deep
- severity
- critical
Medium effort, worth a checkDelete 5 unused symbols and files (52 lines)
Nothing in the graph reaches them, across 2 files; every reader and agent pays to skip them.
- symbols
- 5
- lines
- 52
- safe to delete
- Yes(inferred)
Small effort
Improve what Repowise can see
Add a test coverage report
Without one, Repowise cannot tell tested code from untested code, so every test-related action says “unknown”.
- line coverage
- Unknown, no report
Small effortReview the 46 proposed decisions
None is accepted yet, so nothing can drift from one and agents get no enforced guidance from them.
- proposed
- 46
- accepted
- 0
Medium effort
- Lines of code
- 12.2K
Recent activity
All commitsCommits
- 0d95a81dchore: sync SKILL.md copies [skip ci]3mo ago
- e2c09c9adocs: flatten savings claim to a flat 65% across all surfaces; bump PINNED_REF to v1.9.13mo ago
- dc95e915Caveman Updated Skill + Readme3mo ago
- 686c0cceTeaser; Caveman 23mo ago
- e9cb8435fix(#601): attribute session tokens to the mode active when each message happened3mo ago
- a8846964fix(#597): ship spawn-options.js in caveman-shrink npm package3mo ago
Decisions
- Export spawn options so behavior is unit-testable without rerunning CLIproposed · 2mo ago
- Use shell:false for POSIX spawn to avoid argv quoting surprisesproposed · 2mo ago
- fix(hooks): /caveman arg whitelist + symlinked-parent ~/.claude supportproposed · 2mo ago
- fix(#565): install.ps1 works when piped to iexproposed · 2mo ago
- fix(opencode): real lifecycle hooks + ship caveman-compress commandproposed · 2mo ago
- fix(#601): attribute session tokens to the mode active when each message happenedproposed · 2mo ago
Everything we found
367 open
Every detector's open count by area, each with its worst example. An inventory, not a to-do list: the actions above are the part worth doing.
- worst item is critical severity. Code health231Nested complexity · bin/install.js66 findings · 8.0 deducted
- worst item is medium severity. Refactoring27compress_file+1.2 health, S effort · plugins/caveman/skills/caveman-compress/scripts/compress.py
- worst item is medium severity. Decisions55bin/install.js46 awaiting review · 9 hotspots ungoverned
- worst item is medium severity. Ownership46bin/install.js92% single-owner concentration, 20 commits in 90d
- worst item is low severity. Dead code8evals/measure.pyUnreachable symbol (60 lines)
Composition
Open the graphExplore this codebase
- Docs84 pages across 8 modules
- ChatAsk this codebase a question and get an answer with its sources.
- Files143 files with per-file docs, health and history
- ArchitectureDependency graph, layers, and 4 entry points
- Code healthPer-file scores, 277 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.