jev-rust-review
Claude Code plugin for Rust review: compiler and Clippy facts first, Jev triages semantic risks, Claude reviews and Jev verifies findings.
Overview
jev-rust-review is a Claude Code plugin and MCP server for Rust code review that aims to go beyond the compiler. rustc, Clippy and cargo-semver-checks run first and their output, filtered to changed lines and respecting the project's lint policy, is treated as fact and reported once. TypeSafe's Jev and Claude then focus on what tools do not report, such as a select! branch losing data on cancellation, a check and act under two separate locks, or a length guard removed from a caller in another file.
A Rust context collector feeds Jev triage with one fan-out request per changed unit; flagged unit-dimension pairs, minus anything a tool already reported, go to Claude, which reviews every changed unit with flagged ones first. Candidate findings are then verified and classed as tool-reported, supported, refuted or insufficient context, producing a short, high-precision review. Miri is suggested, never run, when unsafe code changes. The project is v0.1, builds from source on first launch, and is independent of TypeSafe and Anthropic.
What the decision model decides
| Type | Question | Answers |
|---|---|---|
| Yes / no | Does this changed unit carry risk on this dimension (e.g. cancellation safety, locking)?(inferred) | yes · no |
| Choice | Is this candidate finding supported, refuted or lacking context? | tool_reported · supported · refuted · insufficient_context |
How it works
- Run rustc, Clippy and semver checks on changed lines
- Collect Rust context and triage units with Jev
- Drop flags already reported by tools
- Claude reviews units, flagged first
- Verify candidate findings before reporting
Strengths
- No duplication of compiler and Clippy output
- Respects project lint policy
- Verification step for higher precision
Caveats
- v0.1; interfaces may change
- No prebuilt binaries yet
- Requires a TypeSafe key for semantic triage
At a glance
- Type
- Project
- Status
- prototype
- Language
- Rust
- Runtime
- Claude Code plugin with MCP server
- Has code
- yes
- License
- MIT
Use cases and integrations
How curated lists describe it
Rust-aware code review for Claude Code and coding agents, powered by TypeSafe Jev
logicrw/awesome-jev-projectsRust-aware code review for Claude Code and coding agents, powered by TypeSafe Jev.
hellogumbo/awesome-jev
Frequently asked questions
What does Jev decide?
Which changed units carry semantic risk dimensions, and whether candidate findings are supported.
Will it repeat Clippy warnings?
No. Tool-reported defects are never reported again.
Does it run Miri?
No. It suggests Miri when unsafe code changes.
Related projects
Coding-agent hook that asks Jev one yes/no per project rule on every edit or turn and makes the agent fix any violation.
HookbetaTypeScriptAdvisor for coding agents: git and Stop hooks flag test tampering, secrets and skipped hooks, with optional Jev review findings.
HookbetaCLI and agent commands auditing a git diff against YAML standards packs, scoring each hunk with Jev so agents fix only flags.
Agent skillbetaTypeScriptFuzzy linter hook for Claude Code and Codex where Jev checks each written file against team rules and reports violations at edit time.
HookbetaTypeScriptCheap Jev checks beside a Claude coding session: per-function code rules, doc contradiction checks and context retrieval.
ProjectbetaPythonFuzzy linter CLI that checks a code tree against plain-language rules in one markdown file, judged by Jev via a swappable interface.
ProjectbetaTypeScript