check-risk
CLI and GitHub Action that scores a code change's risk from policy rules plus Jev's semantic signals, and suggests checks and reviewers.
Overview
check-risk turns a Git change into an explainable risk score, required checks and reviewer groups, so you understand the risk of an AI-generated (or human) change before merging. Run it in the terminal after a coding session or as a GitHub Action gate on pull requests; no agent plugin is required.
Your policy defines sensitive parts of the repository. Deterministic checks catch changed paths, new dependencies, large diffs and missing test changes, and TypeSafe Jev, through the official SDK, adds semantic signals about the behaviour being changed. Every contribution to the score traces back to a named rule. Output is available as terminal text, JSON, a reusable Action and a TypeScript API. Version 0.1 ships source and an installable CLI archive; npm publication and live Jev validation are still pending.
What the decision model decides
| Type | Question | Answers |
|---|---|---|
| Yes / no | Does this change alter sensitive behaviour defined in the policy (e.g. auth, payments)?(inferred) | yes · no |
How it works
- Read local changes or a base/head comparison
- Apply deterministic policy checks
- Ask Jev for semantic signals about changed behaviour
- Combine into a risk level, score, reasons, checks and reviewers
Strengths
- Every score component traces to a rule
- Works for human and AI code
- Action, CLI and API
Caveats
- v0.1; live Jev validation pending
- Not yet on npm
Getting started
Install the CLI archive from the v0.1 release, export TYPESAFE_API_KEY, and run check-risk in your repository.At a glance
- Type
- CLI
- Status
- prototype
- Language
- TypeScript
- Runtime
- Node CLI + GitHub Action
- Has code
- yes
- License
- MIT
Use cases and integrations
How curated lists describe it
A CLI and GitHub Action that assesses code-change risk using deterministic rules and TypeSafe Jev, recommending checks and reviewers before merge.
hellogumbo/awesome-jev
Frequently asked questions
What does Jev add?
Semantic signals about the behaviour a change affects, on top of deterministic checks.
Does it need an agent?
No, it works on any Git change.
Is it production-ready?
It is v0.1 with live Jev validation pending.
Related projects
Automated PR review: an LLM agent finds issues, a verifier refutes them, and Jev decides how to triage, route and label.
ProjectproductionTypeScriptRust code-review gate asking Jev small typed questions about functions and tests, failing CI only on rules measured 80% right.
CLIproductionRust- CodeOtter★ 3
Self-hosted PR review platform pairing an LLM narrative with a /v1/systemone decision engine for 6 rubric scores and 8 yes/no merge gates.
Servicebeta Coding-agent hook that asks Jev one yes/no per project rule on every edit or turn and makes the agent fix any violation.
HookbetaTypeScriptClaude Code plugin sending small judgments to Jev: rule checks that block edits, verbatim compaction, subagent sizing and routing hints.
PluginbetaPythonAdvisor for coding agents: git and Stop hooks flag test tampering, secrets and skipped hooks, with optional Jev review findings.
Hookbeta