If Jev never writes prose, where does the review text come from?

Jev returns scores, verdicts, and probabilities only. The Worker turns those typed outputs into UI copy, so prose is presentation while decisions stay machine-typed.

When should I use general review versus a job persona?

Use general review for wording, conciseness, structure, metrics, and parseability. Use a stored job persona when a job description plus tags should shape scoring for one resume against one role.

What is the catch with hosted MCP versus local npx mode?

Hosted MCP needs no key but shares a per-minute platform limit. Local npx runs Jev with your own key and skips that limit, with a deterministic mock flag for tests and development.

Why does the tool refuse my PDF upload?

Never send PDF bytes. Extract text first, then pass resume text with an optional lens id or pasted job text; pasted job text wins over a lens id.

What is the stack behind it?

A React and Vite SPA served through @cloudflare/vite-plugin, with a Hono API on a Cloudflare Worker. D1 stores personas, resumes and evaluation runs with an in-memory fallback when the binding is absent, KV backs the unique visitor counter, and Workers Analytics Engine writes to the website_events dataset for pageviews and UI clicks.

How do I run it locally, and what happens without a TypeSafe key?

Run `pnpm install`, copy .dev.vars.example to .dev.vars, optionally set TYPESAFE_API_KEY, then `pnpm dev` and open the Vite URL, usually localhost:5173. Without a key the Worker uses a deterministic mock so the app still runs, and `pnpm dev` applies local D1 migrations first, with the schema also created on the first request if D1 is bound but empty.

Which MCP tools does it expose, and how do I pick a job lens?

list_job_lenses, get_job_lens, suggest_job_lens and review_resume. jobLensId accepts default, a preset id such as swe-staff, or the preset:swe-staff form, and pasted jobText wins over a lens id. Tools return compact JSON with score, findings, suggestions and gaps rather than the full review payload.

CommunityJev Agents & MCP

Jevsume

Resume review that extracts text like a parser would, then runs typed Jev questions through a Cloudflare Worker to compose a score, with general and per-job persona modes.

See highlights

Open github.com

Stars: 3
Language: TypeScript
Published by: Community
Author: unownone
Last verified: Oct 2026

unownone/jevsume

User repository on GitHub

View on GitHub

ATS-friendly resume review powered by Jev (TypeSafe System One). The frontend extracts resume text the way a parser would, then a Cloudflare Worker runs typed JEV questions and composes a JevScore.

Forks: 1
Last pushed: Updated Sep 2026

Repo stats from the GitHub API, cached Oct 2026.Homepage

About this project

ATS-friendly resume review powered by Jev (TypeSafe System One). The frontend extracts resume text the way a parser would, then a Cloudflare Worker runs typed JEV questions and composes a JevScore. General review — wording, conciseness, structure, metrics, ATS parseability. Per-job review — job description + tags become a stored job persona (JEV-native state + questions). One resume is scored against one persona.

Quoted from the project's README. Read the full README.

Tagscommunity awesomejev catalog-import

Works forJev for Classification Jev for Ranking

Used inJev for Software & DevTools

Languages: TypeScript

Highlights

General review plus per-job personas stored as Jev-native state and questions.
MCP tools return compact score, findings, suggestions, and gaps rather than full payloads.
Worker stack on Hono with D1 personas and evaluation runs for later lookup.
Deterministic mock provider lets the app run without a TypeSafe key.

Watch out

Hosted MCP shares a per-minute platform rate limit with the review API; local mode uses your own key instead. Copy reviewed against the repository README on 2026-10-04, not from running it.

Questions about this entry

Short answers drawn from the linked project or page.

More Jev resources like this

jev-test-filter

Diff-aware test selector that sends the git diff to Jev with one question per test, then prints the native filter flags and hands them to vitest, jest, node, bun, Playwright, cargo, or go test.

Jev Agents & MCP#community#jevbest#catalog-import

Community

mizchi/jev-test-filterTypeScript

jev-guard

Claude Code plugin that scores pending tool calls and finished answers with Jev noul probabilities, denying the call with the correction attached when a rule threshold is met.

Jev Agents & MCP#community#awesomejev#catalog-import

Community

muratcakmak/jev-guardTypeScript

pijev

Terminal coding agent on Pi where Jev shortlists skills, ranks retrieved excerpts, and triages tool failures while the coding model handles reasoning and edits.

Jev Agents & MCP#community#awesomejev#catalog-import

Community

tonyzdev/pijevTypeScript

jev-mcp

MCP server exposing Jev Choice, Score, Noul, batched ask, file-reading triage, and model listing as typed tools with full probability distributions.

Jev Agents & MCP#community#jevbest#catalog-import

Community

rashedInt32/jev-mcpTypeScript

typesafe-mod

A Claude Code function hook that sends AskUserQuestion dialogs to Jev and draws a probability per option above the dialog, so you still choose while seeing how confident the model is.

Jev Agents & MCP#community#jevbest#catalog-import

Community

BeLazy167/typesafe-modTypeScript

JevPromptCoach

Claude Code plugin that scores your prompts against seven coding-agent habits on demand rather than on every keystroke, with a backfill so the first report covers prompts you have already written.

Jev Agents & MCP#community#awesomejev#catalog-import

Community

CrowdLinker/JevPromptCoachTypeScript

Back to all resources

From the Jev community

Posts from builders shipping with Jev right now.

Follow @typesafeai