# AI use cases for Jev Review (devagrawal09/jev-review) Repo: [devagrawal09/jev-review](https://github.com/devagrawal09/jev-review) | Use cases: 4 | Page: https://dedupedai.com/use-cases/devagrawal09-jev-review/ A small TypeScript code-review workflow built on TypeSafe Jev that reviews a Git diff or scans a codebase and shows the findings in a local dashboard. This file holds every use case of this repo. Each one starts with a "## " heading and ends with its URL. - Review uncommitted Git changes before committing: https://dedupedai.com/use-cases/devagrawal09-jev-review/review-git-diff-before-commit/ - Scan a whole package for review risks: https://dedupedai.com/use-cases/devagrawal09-jev-review/scan-codebase-for-review-risks/ - Feed a diff review as JSON to a coding agent: https://dedupedai.com/use-cases/devagrawal09-jev-review/pipe-review-json-to-coding-agent/ - Adapt the staged review workflow for your team: https://dedupedai.com/use-cases/devagrawal09-jev-review/extend-staged-review-workflow/ ## How to review uncommitted Git changes before committing with Jev Review Repo: [devagrawal09/jev-review](https://github.com/devagrawal09/jev-review) | Difficulty: Intermediate | Time: 30 minutes | Basis: From the README Get a staged review of your current diff and read the findings in a local dashboard. **Who it is for:** A developer with uncommitted changes in a Git repository who wants a second pass on correctness, security, reliability, compatibility, and test coverage before committing. **The problem:** Diffs are easy to skim past, and it is hard to tell which risks in a change deserve a closer look. **The outcome:** A saved review of the current diff, with findings listed in the dashboard at http://127.0.0.1:4317 as review prompts to check. ### Steps 1. Install Node.js 24 or newer and Git, then create a TypeSafe API key at https://console.typesafe.ai/settings/keys. 2. Run `npm install` in the Jev Review folder. 3. Run `cp .env.example .env`, then add TYPESAFE_API_KEY to the .env file. 4. Run `npm run review:changes:save -- /path/to/git/repository`. 5. Run `npm run dashboard` and open http://127.0.0.1:4317 in a browser. ### Prompt to paste into your agent ```text In the jev-review repo, run `npm run review:changes:save -- ` and then `npm run dashboard`. Report the saved review's findings grouped by category. Verify one finding by reading the cited diff hunk and state whether it looks like a real issue or a false positive. ``` ### Caveat The README says findings are review prompts, not proof of a defect, so each one still needs a check against the code. Risk, credentials: The TypeSafe API key goes in the local .env file, so keep that file out of commits and never paste the key into an agent prompt. 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/devagrawal09-jev-review/review-git-diff-before-commit/ ## How to scan a whole package for review risks with Jev Review Repo: [devagrawal09/jev-review](https://github.com/devagrawal09/jev-review) | Difficulty: Intermediate | Time: 30 minutes | Basis: From the README Run a complete codebase scan on a repository or package and browse the results in the dashboard. **Who it is for:** A maintainer or tech lead inheriting or auditing a package who needs a map of where the riskiest code sits. **The problem:** Reading every source file to find weak spots takes too long, and a diff review does not cover code that has not changed recently. **The outcome:** A saved complete codebase scan viewable in the dashboard, with large reports in collapsible sections to help choose where to read first. ### Steps 1. Install Node.js 24 or newer and Git, then add a TypeSafe API key to .env as described in the Quick Start. 2. Run `npm install` in the Jev Review folder. 3. Run `npm run review:codebase:save -- /path/to/git/repository-or-package`. 4. Run `npm run dashboard` and open http://127.0.0.1:4317 in a browser. 5. Open the collapsible sections for the highest-severity areas first and check each finding against the source. ### Prompt to paste into your agent ```text In the jev-review repo, run `npm run review:codebase:save -- ` and then `npm run dashboard`. Produce a short list of the top five areas flagged, each with its file and the reason given. Verify one area by opening the cited file and confirm the flagged code is actually present. ``` ### Caveat The scan covers every non-ignored source file in the scope, so point it only at code you are allowed to submit to the TypeSafe API. 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/devagrawal09-jev-review/scan-codebase-for-review-risks/ ## How to feed a diff review as JSON to a coding agent with Jev Review Repo: [devagrawal09/jev-review](https://github.com/devagrawal09/jev-review) | Difficulty: Intermediate | Time: 30 minutes | Basis: From the README Print the change review as JSON so an agent can triage the findings against the code. **Who it is for:** A developer using an AI coding agent who wants the agent to act on review findings rather than read them by hand. **The problem:** Agents need structured input, and a plain-text review is hard to parse reliably. **The outcome:** A JSON review of the current diff that the agent can use to decide which findings to fix, dismiss, or investigate. ### Steps 1. Install Node.js 24 or newer and Git, then add TYPESAFE_API_KEY to .env. 2. Run `npm install` in the Jev Review folder. 3. Run `npm run review:changes -- /path/to/git/repository` and save the printed JSON to a file. (inferred, check the README) 4. Give the JSON file to your coding agent with instructions to check each finding against the cited hunk. ### Prompt to paste into your agent ```text In the jev-review repo, run `npm run review:changes -- ` and capture the JSON output. For each finding, open the cited diff hunk in the target repository and label it as confirmed, unclear, or false positive. Verify the labels by re-reading the code before you change anything. ``` ### Caveat The README calls findings review prompts, so an agent should confirm each one in the code and not apply fixes automatically. 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/devagrawal09-jev-review/pipe-review-json-to-coding-agent/ ## How to adapt the staged review workflow for your team with Jev Review Repo: [devagrawal09/jev-review](https://github.com/devagrawal09/jev-review) | Difficulty: Advanced | Time: 45 minutes | Basis: From the README Change the review layers or policy in code and confirm the dependency checks still pass. **Who it is for:** A developer on a team that wants to change which judgments run or how findings are routed. **The problem:** A generic review pass may not match your team's policy or the kinds of mistakes your code makes. **The outcome:** A modified review workflow that still passes the project's typecheck, dependency-flow check, and dashboard syntax check. ### Steps 1. Install Node.js 24 or newer and Git, then add TYPESAFE_API_KEY to .env. 2. Run `npm install` in the Jev Review folder. 3. Edit the workflow files under src/review/ and the policy in src/domain/config.ts to match your team's rules. (inferred, check the README) 4. Run `npm run check` to typecheck, verify dependency flow, and syntax-check the dashboard client. 5. Run `npm run review:changes:save -- /path/to/git/repository` to confirm the modified workflow produces a saved review. ### Prompt to paste into your agent ```text In the jev-review repo, change one threshold in src/domain/config.ts, then run `npm run check`. Report whether typecheck, the dependency check, and the dashboard syntax check pass. Verify that no import now points upward from a lower layer. ``` ### Caveat The README describes this as an experiment that does not yet include compiler diagnostics, static analyzers, or repository indexing, so expect gaps in coverage. 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/devagrawal09-jev-review/extend-staged-review-workflow/