The validated health score Code Climate left behind.
codeclimate.com moved on to AI consulting and spun its quality product out to qlty.sh as a linter. repowise reclaims the validated single health score and the churn x complexity maintainability map they pioneered, plus a wiki, decisions, and agent-native MCP, all open source and self-hostable.
Code Climate proved that a single maintainability grade and a churn-versus-complexity map could focus a team on the code that matters. Then they moved on, and that validated score and hotspot map went unowned.
repowise reclaims exactly that ground: a defect-validated single health score and a churn x complexity maintainability map, kept current and reproducible on your own repo, and bundled into a wider intelligence layer of wiki, decisions, git intelligence, and agent-native MCP.
Which one is right for you?
Choose repowise if
- You want the single maintainability grade and churn x complexity map Code Climate pioneered, still maintained
- You want a health score that is defect-validated and reproducible on your own repo, not just graded
- You want health plus an auto-generated wiki, architectural decisions, and agent-native MCP in one tool
- You want everything open source and self-hostable, with every heuristic inspectable
- You want to give AI coding agents real codebase context
Choose Code Climate / qlty if
- You mainly want a linter: 70+ linters, formatters, and analyzers out of the box (qlty)
- You want a free CLI that aggregates lint and format runs fast across a monorepo (qlty)
- You want one-click auto-format and diff-scoped lint feedback on pull requests (qlty)
- You are buying AI-transformation consulting rather than a code-quality product (codeclimate.com)
repowise vs Code Climate / qlty
| Capability | repowise | Code Climate / qlty |
|---|---|---|
| Single deterministic code-health / maintainability scoreCode Climate's grade is vacated; qlty leads as a linter | Included | Partial support |
| Defect-validated, reproducible on your repo | Included | Not included |
| Three co-equal signals (defect, maintainability, performance)Code Climate blended these into one grade | Included | Not included |
| No account or API key for the first scanrepowise: pip install repowise, zero setup; qlty's CLI is also key-free | Included | Partial support |
| Churn x complexity maintainability mapCode Climate originated it; no longer headlined | Included | Partial support |
| 70+ linters, formatters, and analyzersqlty | Not included | Included |
| Free CLI lint / format aggregationqlty | Not included | Included |
| Hotspots, ownership, and bus factor | Included | Partial support |
| Auto-generated wiki and documentation | Included | Not included |
| Architectural decision records | Included | Not included |
| Agent-native MCP context (10 tools) | Included | Not included |
| AI code provenance (agent attribution) | Included | Not included |
| Dead code detectionqlty flags duplication and some defects | Included | Partial support |
| Open source and self-hostableqlty CLI is open source; Code Climate is sales-led | Included | Partial support |
Self-assessed against publicly documented features as of June 2026. A dash means partial or limited support. Vendor capabilities change, so please verify against Code Climate / qlty's current docs before deciding.
Reclaim the score, and get the layer around it.
The single grade and churn x complexity map Code Climate pioneered, kept validated, plus the context layer it never shipped.
A validated single grade, not an abandoned one
repowise reclaims the single maintainability grade and the churn x complexity hotspot map Code Climate originated, and grounds them in real defects. The validation benchmark runs on your own repo, so you can confirm the score finds your bugs.
- Cross-project ROC AUC 0.74, up to 0.90 per repo
- 2.3x more defects under a fixed review budget in our open 21-repo benchmark
- 21 deterministic biomarkers, zero LLM, under 30 seconds
- Churn x complexity maintainability map for volatile-and-complex files
Health, docs, decisions, and agent context together
qlty is a fast, broad linter; Code Climate moved to consulting. repowise puts the validated health score alongside an auto-generated wiki, architectural decision archaeology, git intelligence, agent provenance, and ten MCP tools, so the same index serves your quality goals and your AI agents.
- Auto-generated wiki, rebuilt on every commit
- Architectural decisions mined from eight sources
- Agent provenance: how much of your code AI wrote, and whether it is healthy
- MCP tools for Claude Code, Cursor, Cline, and Codex
Inspect every heuristic, run it offline
repowise is free and open source to self-host under AGPL-3.0, with hosted tiers priced per repo and per seat. Every biomarker and weight is public, so the grade is auditable rather than a black box.
- AGPL-3.0: inspect, fork, self-host the full platform
- Bring your own LLM key or run fully offline
- Zero telemetry, code never leaves your infrastructure
- Commercial license available when you need it
What Code Climate blended, repowise keeps separate.
Code Climate's maintainability grade folded readability, complexity, and change-cost together into one number. repowise scores three co-equal pillars instead, so a dip in one never quietly drags down another.
Defect risk
The primary 1-to-10 score, from 21 markers validated against real defects (ROC AUC 0.74). This is the number to watch for bug risk.
Maintainability
8 markers for readability and change-cost smells that do not, on their own, predict bugs, reported as its own score rather than blended into defect risk.
Performance
20 markers for static I/O-in-loop and N+1 risk shapes, a third co-equal pillar, high-precision and deliberately narrow in scope.
None of this needs an account or an LLM key to run. pip install repowise, then repowise init --yes --no-prose builds the dependency graph, git history, and all three pillars in one pass, before you decide whether to add the wiki's prose narrative at all.
The honest version
qlty is a genuinely strong linter, and there are places it leads today. It ships 70+ linters, formatters, and security analyzers out of the box with a free CLI, and it aggregates lint and format runs fast across a monorepo with diff-scoped feedback on pull requests. If broad linter coverage and a free CLI are what you need, qlty is a strong choice, and codeclimate.com may be the right partner if you are buying AI-transformation consulting rather than a code-quality product. repowise wins when you want the validated single health score and churn x complexity map they pioneered, kept current, with a full context layer and agent-native access on top.
What repowise costs.
The repowise core is open source under AGPL-3.0 and free to self-host, with every heuristic public. Hosted tiers: Free for public repos, Pro at $15 per month, Teams at $60 per month, and custom enterprise licensing when you need it. Code Climate / qlty's pricing changes, so verify it on their site, then compare it with a tier you can read line by line.
Questions, answered
Is repowise a good Code Climate Quality alternative?
Yes, if you relied on Code Climate's single maintainability grade and its churn-versus-complexity hotspot map. codeclimate.com has moved on to AI-transformation consulting and spun the quality product out to qlty.sh, which now leads as a linter. repowise reclaims the validated single health score and the churn x complexity maintainability map, and adds an auto-generated wiki, architectural decisions, and ten MCP tools for AI agents.
What happened to Code Climate Quality?
The team moved on. codeclimate.com repositioned as an AI-transformation consultancy, and the code-quality product was spun out to qlty.sh, where it is positioned as a linter rather than a maintainability grade. We kept building the validated score and the churn x complexity map that Code Climate originated.
Does repowise have a single code-health score like Code Climate's maintainability grade?
Yes. repowise scores every file 1 to 10 from 21 deterministic biomarkers (complexity, nesting, cohesion, clones, change entropy, co-change scatter, ownership dispersion, prior-defect history, and more), with no LLM, in under 30 seconds on a typical repo. It is the validated descendant of the single grade Code Climate pioneered, and it stays a defect-risk score specifically: maintainability and performance are separate, co-equal scores rather than being folded into it.
Does repowise have a churn versus complexity map?
Yes. repowise ships a churn x complexity maintainability map that surfaces the volatile-and-complex files most likely to hurt you, the same hotspot quadrant Code Climate originated and no longer headlines.
How is repowise's health score validated?
repowise publishes its predictive performance against real defect labels and lets you reproduce it on your own repo: cross-project ROC AUC 0.74 (95% CI 0.68 to 0.79, up to 0.90 per repo), with 2.3x more defects surfaced under a fixed review budget in its open 21-repo benchmark. Every biomarker and weight is open source under AGPL-3.0.
Where is qlty stronger than repowise?
qlty ships 70+ linters, formatters, and security analyzers out of the box with a free CLI, and aggregates lint and format runs fast across a monorepo. If broad linter coverage and a free CLI are your priority, qlty is a strong choice. repowise wins when you want the validated single health score, the churn x complexity map, a context layer, and agent-native access.
Can repowise give AI coding agents codebase context?
Yes, and this is a core difference. repowise exposes the whole index through ten MCP tools (get_overview, get_answer, get_context, get_risk, get_why, and more) so Claude Code, Cursor, Cline, and Codex answer from a real model of your code. Neither Code Climate nor qlty offers an equivalent agent-native context layer.
Does repowise blend maintainability into the health grade, like Code Climate did?
No, and this is a deliberate difference. repowise reports three co-equal signals instead of one blended grade: defect risk (the 21-marker score above), maintainability (8 markers covering readability and change-cost smells that do not on their own predict bugs), and performance (20 markers covering static I/O-in-loop and N+1 risk shapes). Code Climate's single grade mixed these together; repowise keeps them separate so a maintainability dip cannot quietly move the defect-risk number.
Do I need an API key to run repowise's scoring?
No. pip install repowise, then repowise init --yes --no-prose builds the dependency graph, git history, and the full three-pillar health score with zero LLM calls. Neither Code Climate's dashboard nor qlty's CLI requires an LLM key either for their core checks, but repowise's defect-validated score and its churn x complexity map both come free of any account or cloud step.
The defect-validated score, in depth.
21 markers, the ROC AUC methodology, and how to read a worklist.
Hotspots, ownership, coupling, bus factor.
Where a validated, reproducible score fits among refactoring tools.
Why nesting depth catches what a flat complexity count misses.
Health, ownership, and AI-debt for leaders.