# AI use cases for Jev Lint (mizchi/jev-lint) Repo: [mizchi/jev-lint](https://github.com/mizchi/jev-lint) | Use cases: 5 | Page: https://dedupedai.com/use-cases/mizchi-jev-lint/ jev-lint is a command-line lint tool that uses an AI classifier to flag code, commits, shell scripts and documents whose claims contradict what they actually do. This file holds every use case of this repo. Each one starts with a "## " heading and ends with its URL. - Find doc comments that contradict their TypeScript code: https://dedupedai.com/use-cases/mizchi-jev-lint/catch-comments-that-lie-in-typescript/ - Review only the lines a pull request changed: https://dedupedai.com/use-cases/mizchi-jev-lint/review-only-changed-lines-before-merge/ - Check commit messages and staged changes against AGENTS.md: https://dedupedai.com/use-cases/mizchi-jev-lint/check-commit-messages-against-agents-md/ - Audit shell scripts for hidden effects on the machine: https://dedupedai.com/use-cases/mizchi-jev-lint/audit-shell-scripts-for-hidden-machine-effects/ - Let a coding agent install the skill and run Jev Lint: https://dedupedai.com/use-cases/mizchi-jev-lint/let-coding-agent-run-jev-lint/ ## How to find doc comments that contradict their TypeScript code with Jev Lint Repo: [mizchi/jev-lint](https://github.com/mizchi/jev-lint) | Difficulty: Intermediate | Time: 30 minutes | Basis: From the README Flag functions whose comments promise behaviour the body does not deliver. **Who it is for:** A TypeScript developer whose doc comments and return types have drifted from the implementation after refactors, and whom the compiler and ESLint do not warn about. **The problem:** Comments such as 'returns null when not found' stay in place while the body throws, and nothing in the toolchain reports the mismatch. **The outcome:** A list of findings with a probability score and cutoff for each suspect block, which you review and fix by hand. ### Steps 1. `export TYPESAFE_API_KEY=...` 2. `npx -y jev-lint check examples` 3. `npx jev-lint check src --dry-run` 4. `npx jev-lint check src` ### Prompt to paste into your agent ```text In the repository , set TYPESAFE_API_KEY, run `npx jev-lint check src --dry-run` to see the planned requests and price, then run `npx jev-lint check src`. Report each finding with file, line, rule id and score, and do not edit code until a human confirms the finding is real. ``` ### Caveat Findings are candidates, not verdicts: the README reports about one in five was wrong on this repository, so each one needs a human check. Disclaimer: Third-party repos we don't own or vouch for. AI-generated use cases can be wrong: review everything and run at your own risk. Full text: https://dedupedai.com/disclaimer/ URL: https://dedupedai.com/use-cases/mizchi-jev-lint/catch-comments-that-lie-in-typescript/ ## How to review only the lines a pull request changed with Jev Lint Repo: [mizchi/jev-lint](https://github.com/mizchi/jev-lint) | Difficulty: Advanced | Time: 45 minutes | Basis: From the README Run Jev Lint on a diff so findings land on new code and cost less. **Who it is for:** A maintainer reviewing pull requests on a busy repo who wants AI-judged feedback on the touched lines without linting the whole codebase. **The problem:** Checking every file on each pull request is slow and costly, while most findings concentrate in changed code. **The outcome:** Annotated findings on the pull request's changed lines, with the rest of the codebase left out of the run. ### Steps 1. `npm install --save-dev jev-lint` 2. `npx jev-lint init` 3. `npx jev-lint review --base main` (inferred, check the README) 4. `review --base "$GITHUB_BASE_REF" --format github` ### Prompt to paste into your agent ```text In the repository on a feature branch, with TYPESAFE_API_KEY set, run `npx jev-lint review --base main` and list each finding with file, line, rule and score. Verify that every finding points at a line the diff actually changed. ``` ### Caveat The README says CI can run a review, but no shipped rule blocks a merge until you raise a rule to severity: error, so the check is advisory by default. Disclaimer: Third-party repos we don't own or vouch for. AI-generated use cases can be wrong: review everything and run at your own risk. Full text: https://dedupedai.com/disclaimer/ URL: https://dedupedai.com/use-cases/mizchi-jev-lint/review-only-changed-lines-before-merge/ ## How to check commit messages and staged changes against AGENTS.md with Jev Lint Repo: [mizchi/jev-lint](https://github.com/mizchi/jev-lint) | Difficulty: Intermediate | Time: 30 minutes | Basis: From the README Judge each commit message against its diff and each change against repo instructions. **Who it is for:** A developer or agent working in a repository that has an AGENTS.md file and wants commits and staged edits checked against those written rules before they are pushed. **The problem:** Commit messages can describe something other than the diff, and changes can break instructions the repository wrote for itself. **The outcome:** A per-commit report of message-versus-diff and change-versus-instruction findings, plus a check of staged work before the commit is made. ### Steps 1. `export TYPESAFE_API_KEY=...` 2. `npx jev-lint commits --base main` (inferred, check the README) 3. `npx jev-lint commits --staged` (inferred, check the README) ### Prompt to paste into your agent ```text In the repository , with TYPESAFE_API_KEY set, run `npx jev-lint commits --base main` and then `npx jev-lint commits --staged`. Report each finding that cites an AGENTS.md instruction or a commit message mismatch, and confirm each cited instruction exists in the file. ``` ### Caveat The README says only instructions a diff can be held against are judged, so vague guidance such as 'develop test-first' will not produce findings. Disclaimer: Third-party repos we don't own or vouch for. AI-generated use cases can be wrong: review everything and run at your own risk. Full text: https://dedupedai.com/disclaimer/ URL: https://dedupedai.com/use-cases/mizchi-jev-lint/check-commit-messages-against-agents-md/ ## How to audit shell scripts for hidden effects on the machine with Jev Lint Repo: [mizchi/jev-lint](https://github.com/mizchi/jev-lint) | Difficulty: Intermediate | Time: 30 minutes | Basis: From the README Flag scripts that download and run code, read secrets or install persistence. **Who it is for:** A DevOps engineer or security-minded reviewer inheriting shell scripts from an unknown source, who needs to see what they do to the host. **The problem:** A script's effects on the machine, such as running downloaded code or installing persistence, are often invisible to its reader. **The outcome:** A ranked list of shell findings with scores, to guide a manual security read of the flagged lines. ### Steps 1. `npx -y jev-lint check examples` 2. `npx jev-lint check scripts --dry-run` (inferred, check the README) 3. `npx jev-lint check scripts` (inferred, check the README) 4. `npx jev-lint run shell/ scripts` (inferred, check the README) ### Prompt to paste into your agent ```text In the repository , with TYPESAFE_API_KEY set, run `npx jev-lint check scripts --dry-run` to confirm what would be sent, then `npx jev-lint check scripts`. List findings on .sh, .bash or .zsh files with their rule ids, and flag any that a human should read before running the script. ``` ### Caveat The shell rules are calibrated to the same bar as the rest but not promised as first tier, so treat their findings as leads for review. Risk, remote script: Scripts downloaded from unknown sources should be audited before execution, and this tool only flags candidates; it does not prove a script is safe. Disclaimer: Third-party repos we don't own or vouch for. AI-generated use cases can be wrong: review everything and run at your own risk. Full text: https://dedupedai.com/disclaimer/ URL: https://dedupedai.com/use-cases/mizchi-jev-lint/audit-shell-scripts-for-hidden-machine-effects/ ## How to let a coding agent install the skill and run Jev Lint with Jev Lint Repo: [mizchi/jev-lint](https://github.com/mizchi/jev-lint) | Difficulty: Intermediate | Time: 30 minutes | Basis: From the README Install the Jev Lint skill so an agent can run checks and write project rules. **Who it is for:** A developer using Claude Code, Codex, Cursor or Gemini CLI who wants the agent to lint its own changes and draft rules for the project. **The problem:** Agents often lack a repeatable way to check their output against the repository's claims and instructions. **The outcome:** An agent that installs the Jev Lint skill, runs the right command for a target, and reports findings it can verify. ### Steps 1. `npx skills add mizchi/jev-lint --skill jev-lint` 2. `export TYPESAFE_API_KEY=...` 3. `npx jev-lint check src --dry-run` 4. `npx jev-lint review --base main` (inferred, check the README) ### Prompt to paste into your agent ```text In the repository , install the skill with `npx skills add mizchi/jev-lint --skill jev-lint`, then run `npx jev-lint check src --dry-run` and report the planned request count and price. Only after a human approves, run `npx jev-lint review --base main` and list findings with file, line and score. ``` ### Caveat The README warns the repository also ships jev-lint-repo, a maintainer skill, so the install must use --skill Jev Lint to avoid the wrong one. Disclaimer: Third-party repos we don't own or vouch for. AI-generated use cases can be wrong: review everything and run at your own risk. Full text: https://dedupedai.com/disclaimer/ URL: https://dedupedai.com/use-cases/mizchi-jev-lint/let-coding-agent-run-jev-lint/