jev-review
Staged code-review workflow for JS/TS where Jev screens risk, picks evidence, classifies mechanism, scores severity and routes reviewers.
Overview
Jev Review is a small code-review workflow built on TypeSafe Jev that can review a Git diff or scan a whole codebase, then show results in a local dashboard. Orchestration stays in code and Jev makes bounded judgments in stages: a Noul risk matrix across correctness, security, reliability, compatibility and test coverage; Choice and Score file profiles; a Choice selecting concrete diff hunks or source regions as evidence; a Choice classifying the mechanism; a Score for severity; and a conditional Choice routing to a reviewer.
It uses changed or related tests as context for test-gap judgments, structured hints, counterexamples and explicit decision boundaries, with thresholds and policy applied in code and no generative model involved. The dashboard binds to localhost and never serves environment files. The code is layered with a dependency check that fails on upward imports or cycles. The author calls it an experiment that does not yet integrate compiler diagnostics, static analysers or repository indexing.
What the decision model decides
| Type | Question | Answers |
|---|---|---|
| Yes / no | Does this change carry a correctness / security / reliability / compatibility / test risk? | yes · no |
| Choice | Which hunk or region is the evidence for this risk? | — |
| Score | How severe is this finding? | — |
| Choice | Which reviewer should look at this? | — |
How it works
- Collect a Git diff or source files
- Run the Noul risk matrix
- Profile files and select evidence regions
- Classify mechanism and score severity
- Route to a reviewer and show the dashboard
Strengths
- No generative model; bounded judgments only
- Evidence selected before scoring
- Clean, enforced architecture
Caveats
- Experimental; no compiler or static-analysis integration
- JavaScript and TypeScript focus
- Requires Node 24+ and a TypeSafe key
Getting started
npm install, cp .env.example .env (add TYPESAFE_API_KEY), npm run review:changes:save -- /path/to/repo, then npm run dashboardAt a glance
- Type
- App
- Status
- prototype
- Language
- TypeScript
- Runtime
- Node CLI plus local dashboard
- Has code
- yes
- License
- MIT
Use cases and integrations
How curated lists describe it
Software engineering: staged code-review workflow and local dashboard where Jev gates each review stage before a change advances.
yibie/awesome-jevStaged code-review workflow and local dashboard driven by focused Jev calls.
AnotiaWang/awesome-decision-modelsReviews a Git diff or a codebase in stages and displays review leads in a local dashboard.
logicrw/awesome-jev-projectsA staged code-review workflow and local dashboard built with TypeSafe Jev.
wh000wh000/awesome-jev-liveStaged code-review workflow with a local dashboard.
cobanov/awesome-jevA staged code-review workflow and local dashboard built with TypeSafe Jev.
hellogumbo/awesome-jevStaged code-review workflow for JavaScript and TypeScript with a local dashboard. Jev screens correctness, security, reliability, compatibility, and test risk, then scores severity and suggests a reviewer, with no generative model involved.
MrJev/awesome-jev
Frequently asked questions
What does Jev decide?
Risk flags, evidence regions, mechanism, severity and reviewer routing for code changes.
Does it generate review comments?
No. There is no generative model; results are structured judgments.
Can it scan a whole codebase?
Yes, in addition to reviewing the current diff.
Related projects
CLI that scans source, build and CI files with Jev for hidden, deceptive or data-stealing behavior before you run unfamiliar code.
CLIbetaTypeScriptUnix-style tools where Jev learns a repo's decision norms and judges each node of a code lead as a bug, lead or clean.
ProjectbetaJavaScriptRust code and security reviewer that asks Jev staged questions about diffs and posts findings with severity and owner routing.
ProjectbetaRustSemantic code linter that reads each method with its callers and callees and asks decision models about defects, kind, line and severity.
CLIbetaTypeScriptProof-of-concept scanner that asks Jev hundreds of bug questions per Solidity file and writes a heat map of likely vulnerabilities.
ProjectprototypePythonRust CLI that classifies Git commit diffs with Jev into bug fixes, security fixes with CWEs, and change types.
CLIbetaRust