New: 98,000+ skills indexed from GitHub repos with 100+ stars · every skill safety-checkedBrowse top skillsTop skills
Esc to closeBrowse all skills
Submit a skillBrowse skills
Submit a skillBrowse skills

Agent skill

Resume To Job Screen

by mrmps in mrmps/classifier-dev

Score resumes against a job description one explicit competency at a time, with a keyless classification API, ordered labels (no evidence, mentions, demonstrated, strong) and an expected value taken…

MITAuto-check passedBusiness, Finance & HR

Install Resume To Job Screen

skills CLI
$ npx skills add mrmps/classifier-dev --skill resume-to-job-screen -a claude-code

Project install by default; add -g for ~/.claude/skills/.

GitHub CLI
$ gh skill install mrmps/classifier-dev resume-to-job-screen --agent claude-code

Project scope by default; add --scope user for a personal install. Needs GitHub CLI 2.90.0 or later (public preview).

Manual copy
$ git clone --depth 1 https://github.com/mrmps/classifier-dev.git skills-src && mkdir -p .claude/skills && cp -r skills-src/skills/resume-to-job-screen .claude/skills/resume-to-job-screen && rm -rf skills-src

Use ~/.claude/skills/ instead of .claude/skills for a personal install. The folder must contain SKILL.md.

Claude Code skills documentation · loads skills from .claude/skills/

Facts

Skill name
resume-to-job-screen
GitHub stars
424
Token cost
~1.5k tokens
SKILL.md length
677 words
Files
1
Skills in repo
21
Repo updated
First seen
Licence
MIT

At a glance

Score resumes against a job description one explicit competency at a time, with a keyless classification API, ordered labels (no evidence, mentions, demonstrated, strong) and an expected value taken…

  • Works in 4 steps: write competencies from the JD, not a… → one call per competency, ordered labels → score the expected value, not the… → …
  • Tasks that involve Recruiting and HR
  • SKILL.md covers What this must not be used for, Step 1 — write competencies…, Step 2 — one call per… and Step 3 — score the expected…, plus 3 more sections
  • Calls curl; reaches classifier.dev

What it does

Resume To Job Screen is an agent skill from mrmps/classifier-dev. Score resumes against a job description one explicit competency at a time, with a keyless classification API, ordered labels (no evidence, mentions, demonstrated, strong) and an expected value taken from the returned score distribution, producing a weighted ranked shortlist with every number shown. Use on "screen these CVs", "rank these applicants against the JD", "who is worth an interview", "build a shortlist". A reading aid only: it never decides, and it scores evidence in the text, never who the candidate is.

Its SKILL.md is about 1.5k tokens, which your agent loads only when the skill is triggered. It is a single SKILL.md file with no bundled scripts.

It sits in Business, Finance & HR, covering Recruiting and HR. The repository describes itself as: Zero-shot text classification over plain HTTP — no API key, no account. One Cloudflare Worker, a CLI, and an MCP server. https://classifier.dev. The licence is MIT.

When your agent uses it

  • Tasks that involve Recruiting and HR

Example prompts

  • “screen these CVs”
  • “rank these applicants against the JD”
  • “who is worth an interview”
  • “/resume-to-job-screen”

Requirements

  • Python 3

Workflow steps

4 steps, taken from the step headings in SKILL.md.

  1. write competencies from the JD, not a summary of it
  2. one call per competency, ordered labels
  3. score the expected value, not the winning label
  4. the shortlist, with the working shown

What it can do on your machine

Read from SKILL.md and the folder at commit 629df75. It shows what the files ask for, not the result of running them.

  • Tool permissions

    Pre-approves nothing: there is no allowed-tools line, so your agent's usual permission prompts apply.

    From allowed-tools in the SKILL.md frontmatter.

  • Runs code

    Shell commands in SKILL.md call:

    • curl

    From the folder's file list and the shell code blocks in SKILL.md.

  • Network

    Hosts in commands or code, which the agent is likely to contact:

    • classifier.dev

    From URLs in SKILL.md, links to its own repository left out.

  • Credentials

    Names no API keys, tokens, secrets or passwords.

    From names ending in _API_KEY, _TOKEN, _SECRET, _KEY or _PASSWORD in SKILL.md.

Context cost

Resume To Job Screen loads about 1.5k tokens when it runs. Until then it costs about 135 tokens; SKILL.md has 677 words of instructions outside code blocks.

Always · name and description, kept in context so the agent knows when to use it
~135
When it runs · the whole SKILL.md, loaded when a task matches
~1.5k

Estimates: characters ÷ 4, the usual rule of thumb; real counts depend on the model's tokenizer. Scripts and assets cost tokens only if the agent reads them.

Safety

Auto-check passed

The automated check found no risky patterns in SKILL.md.

Automated static check — not a guarantee. Review scripts before installing. It scans the text of SKILL.md for risky patterns (piping downloads into a shell, reading credential files, hidden Unicode, destructive commands); files beside SKILL.md are not scanned.

SKILL.md

The full file from mrmps/classifier-dev at commit 629df75, republished under its MIT licence (© mrmps). 677 words, ~1,499 tokens.

Download SKILL.mdSave it as .claude/skills/resume-to-job-screen/SKILL.md (or your agent's skills folder).
name
resume-to-job-screen
description
Score resumes against a job description one explicit competency at a time, with a keyless classification API, ordered labels (no evidence, mentions, demonstrated, strong) and an expected value taken from the returned score distribution, producing a weighted ranked shortlist with every number shown. Use on "screen these CVs", "rank these applicants against the JD", "who is worth an interview", "build a shortlist". A reading aid only: it never decides, and it scores evidence in the text, never who the candidate is.
license
MIT

Screen resumes against a job description, one competency at a time

classifier.dev picks one of your labels for a text and returns the whole score distribution with it. No key. That gives you a defensible first pass: each competency scored on its own, every number visible, the ranking reproducible. It writes nothing and decides nothing.

What this must not be used for

  • Never classify or infer a protected characteristic: age, sex, race, religion, disability, pregnancy, nationality, marital status, or a proxy for one such as graduation year, name, photo, or a gap in employment. Do not put such a label in labels or a hint in instructions.
  • Never auto-reject. The output ranks and shows evidence; a person reads every resume before anyone is turned down, and decides.
  • Do not use it where local law regulates automated decisions in hiring (the EU AI Act treats hiring as high risk; New York City Local Law 144 requires a bias audit and notice). Confirm what applies before it touches a real applicant.
  • Do not score "culture fit", "communication skills" or "attitude" from a resume. The text does not carry them, so the answer is noise with a number.

Strip names, addresses, dates of birth, photos and school names from the text before sending. You are scoring evidence of work, and nothing else improves the answer.

Step 1 — write competencies from the JD, not a summary of it

Pull three to six competencies out of the JD that a resume could actually show. Each needs a sentence, not a keyword: Python services running in production, not notebooks or scripts beats Python. Attach a weight to each; they sum to 1.

Step 2 — one call per competency, ordered labels

The same four labels for every competency, in order, with a leading digit so the scores map is easy to turn into a number:

curl -s https://classifier.dev/v1/classify \
  -H 'content-type: application/json' \
  -d '{
  "labels": ["0 no evidence in the resume",
             "1 mentions the area without detail",
             "2 did this on the job",
             "3 owned or led this, with scale or outcomes"],
  "instructions": "Score only this competency: designing and operating distributed systems, queues, sharding, on-call. Judge evidence in the resume text, never the person.",
  "inputs": ["Senior engineer, 6 years. Python and Go services on AWS, built a sharded job queue handling 3M jobs a day, on-call rotation owner.",
             "Backend developer, 4 years. Django and Flask internal tools, some Celery, familiar with Kubernetes."]
}'

Real output, with the expected value computed from scores:

label 3  confidence 1.00  scores {0:0.00, 1:0.00, 2:0.00, 3:1.00}  EV 3.00
label 1  confidence 0.56  scores {0:0.32, 1:0.67, 2:0.01, 3:0.00}  EV 0.69

One call per competency, up to 1,000 resumes each. Keep the requests identical apart from instructions: same labels, same order, or the numbers stop comparing.

Step 3 — score the expected value, not the winning label

python
def ev(result):          # the mean of the distribution, 0.00 to 3.00
    return sum(int(label[0]) * s for label, s in result["scores"].items())

total = sum(weight[c] * ev(per_competency[c][i]) for c in competencies)

The argmax label throws away the distribution. The second candidate above sits between 0 and 1 at 0.56 confidence: as a label that is a coin flip, as an expected value, 0.69, a stable "barely anything here". Take the EV and the near-ties stop mattering.

Show full SKILL.md (280 more words)Show less

Step 4 — the shortlist, with the working shown

Five real resumes, three competencies weighted 0.40 production Python, 0.35 distributed systems, 0.25 leading engineers. Three calls, 15 classifications:

cand  python  distrib  leading  total
A     3.00    2.99     2.72     2.93
E     2.94    2.99     2.33     2.80
C     0.00    0.74     2.99     1.01
B     1.58    0.52     0.04     0.82
D     0.25    0.32     0.00     0.21

C is the row that proves the method: an engineering manager, three years out of code, who would have ranked well on any "senior, 11 years" keyword filter and ranks third here because the Python column is 0.00. Hand the table to the hiring manager with the per-competency numbers, not the total alone.

The confidence gate

Per competency, per candidate:

  • 0.9 and above — take the score into the ranking.
  • 0.5 to 0.9 — take it, and mark the cell. The EV is usually still right; the label is what is uncertain.
  • Under 0.5 — do not let that cell move the ranking. A person reads that section of the resume and scores it, or you drop the competency as unscorable from a resume.

Anyone near the cut line is read by a person anyway. The ranking orders the reading; it does not replace it.

Pitfalls

  • A resume is a claim. The model scores what is written, so someone who writes well outscores someone who did more and wrote less. That is exactly why this ranks a reading order rather than deciding.
  • Keep the raw output. Save every request and response. If a candidate asks why they were not progressed you need the numbers, and under some hiring laws you must be able to produce them.
  • Re-run the whole batch when you change a label. Scores from different label sets are not comparable, and a mixed table is worse than no table.
  • Limits per IP: 3,000 classifications a minute, 20,000 a day. A 429 carries Retry-After.

© mrmps, MIT. Rendered from Markdown: HTML in the file is shown as text, images as links, and headings moved down two levels. Raw file

Files

Just SKILL.md in skills/resume-to-job-screen of mrmps/classifier-dev.

Open the folder on GitHubat commit 629df75

Compare with similar skills

Resume To Job Screen next to the 5 skills that share the most tags, products or categories with it. Stars are the repository's; “used in” counts other GitHub owners with a copy.

Resume To Job Screen compared with similar skills
SkillStarsUsed inTokensAuto-checkLicenceRepo updated
Resume To Job Screen this skillmrmps/classifier-dev424—~1.5kAutomated safety check: PassMIT
Get Jobagentenatalie/get-job.skill632—~1.7kAutomated safety check: PassCC-BY-NC-ND-4.0
Resume Reviewerweeelin98/ResumeDom173—~2.4kAutomated safety check: PassNone
Build Resume Portfolio Sitetao943/build-resume-portfolio-site195—~5.8kAutomated safety check: PassNone
Cyber Resume Reviewermubix/cyber-resume-reviewer-skill184—~2.9kAutomated safety check: PassMIT
Repo To Resume TailorSsabby1/repo-to-resume-tailor127—~1.8kAutomated safety check: PassMIT

Similar skills

  • Get Job

    agentenatalie/get-job.skill

    实习.skill / get-job.skill：从岗位调研、简历改写到分轮次面试准备的全流程求职 skill。适合找工作、投实习、校招、秋招、春招、暑期实习、社招、跳槽、转行、跨专业求职、留学生求职，以及产品经理、运营、市场、咨询、AI 产品、AI Coding、数据分析、技术岗等目标岗位准备。

    632 GitHub stars~1.7k tokensUpdated 1 mo ago
    Business, Finance & HRAuto-check passed

  • Resume Reviewer

    weeelin98/ResumeDom

    Build, assess, review, and tailor evidence-backed US-market technology resumes for computer-science interns and new graduates.

    173 GitHub stars~2.4k tokensUpdated 15 days ago
    Business, Finance & HRAuto-check passed

  • Build Resume Portfolio Site

    tao943/build-resume-portfolio-site

    A skill your agent uses when turning resume materials and an optional job description into verified, approved content and a runnable React + Vite resume or portfolio site, or when redesigning an…

    195 GitHub stars~5.8k tokensUpdated 10 days ago
    Business, Finance & HRAuto-check passed

  • Cyber Resume Reviewer

    mubix/cyber-resume-reviewer-skill

    Review, tailor, score, or rewrite IT and cybersecurity resumes.

    184 GitHub stars~2.9k tokensUpdated 20 days ago
    Business, Finance & HRAuto-check passed

  • Repo To Resume Tailor

    Ssabby1/repo-to-resume-tailor

    Analyze a full code repository and generate one resume-ready project description grounded in repository evidence.

    127 GitHub stars~1.8k tokensUpdated 6 mo ago
    Business, Finance & HRAuto-check passed

  • GitHub Project Contributor Finder API Skill

    browser-act/skills

    This skill helps users extract GitHub repository project details and contributor contact information using keywords, stars, and update dates.

    6.1k GitHub starsUsed in 1 repo~1.9k tokens
    Business, Finance & HRAuto-check passed

More from mrmps/classifier-dev

All 21 skills in this repo

  • Bulk Classify

    mrmps/classifier-dev

    Sort many texts into your own categories without reading them, using a keyless HTTP API that returns a calibrated confidence per answer.

    424 GitHub stars~3.1k tokensUpdated 4 days ago
    Auto-check passed

  • Computer Use Action Picker

    mrmps/classifier-dev

    Pick a browser or desktop agent's next action by choosing among the actions actually on screen instead of inventing one.

    424 GitHub stars~1.5k tokensUpdated 4 days ago
    Auto-check passed

  • Content Moderation Gate

    mrmps/classifier-dev

    Check user-generated text against a written policy before it is published.

    424 GitHub stars~1.5k tokensUpdated 4 days ago
    Auto-check passed

  • Context Compaction Keep Or Drop

    mrmps/classifier-dev

    Label each context chunk keep, drop or replace-with-a-pointer and pass the survivors through byte for byte instead of summarising, with key-shaped chunks decided locally and never sent, and a…

    424 GitHub stars~1.5k tokensUpdated 4 days ago
    Auto-check passed

  • Document Intake Routing

    mrmps/classifier-dev

    Label each page of an intake packet with a document type and a page role before extraction runs, so only confident pages reach an extractor and the rest reach a person.

    424 GitHub stars~1.5k tokensUpdated 4 days ago
    Auto-check passed

  • Headline Filter Map Reduce

    mrmps/classifier-dev

    Filter hundreds or thousands of headlines, search results or feed items against a written brief before opening any of them, using a two-stage cascade that spends a fast model on everything and a…

    424 GitHub stars~1.5k tokensUpdated 4 days ago
    Auto-check passed

Categories

Questions about Resume To Job Screen

What does Resume To Job Screen do?

Score resumes against a job description one explicit competency at a time, with a keyless classification API, ordered labels (no evidence, mentions, demonstrated, strong) and an expected value taken…. Resume To Job Screen is an agent skill from mrmps/classifier-dev. Score resumes against a job description one explicit competency at a time, with a keyless classification API, ordered labels (no evidence, mentions, demonstrated, strong) and an expected value taken from the returned score distribution, producing a weighted ranked shortlist with every number shown.

When should I use Resume To Job Screen?

Resume To Job Screen fits situations like: tasks that involve Recruiting and HR.

How do I install Resume To Job Screen in Claude Code?

Run `npx skills add mrmps/classifier-dev --skill resume-to-job-screen -a claude-code`. Or copy the skill folder (skills/resume-to-job-screen in mrmps/classifier-dev) into .claude/skills/resume-to-job-screen in your project. Claude Code loads it when a task matches its description.

How do I install Resume To Job Screen in Codex?

Run `npx skills add mrmps/classifier-dev --skill resume-to-job-screen -a codex`. Or copy the skill folder (skills/resume-to-job-screen in mrmps/classifier-dev) into .agents/skills/resume-to-job-screen in your project. Codex loads it when a task matches its description.

Can I use Resume To Job Screen in Cursor, Gemini CLI or GitHub Copilot?

Cursor, Gemini CLI, GitHub Copilot and OpenCode also load SKILL.md folders. With the skills CLI, run `npx skills add mrmps/classifier-dev --skill resume-to-job-screen -a cursor` (or -a gemini-cli, github-copilot or opencode for the others). To copy it by hand, put the folder in .cursor/skills/resume-to-job-screen, .gemini/skills/resume-to-job-screen, .github/skills/resume-to-job-screen and .opencode/skills/resume-to-job-screen in your project.

What does Resume To Job Screen need to run?

Going by SKILL.md and its folder, Resume To Job Screen needs the command-line tools its instructions call (curl). Our summary lists: Python 3.

Does Resume To Job Screen access the network?

SKILL.md names 1 domain. In commands or code: classifier.dev; the agent is likely to contact it when it follows the instructions. This is read from the text; nothing was executed.

Is Resume To Job Screen safe to install?

Our automated static check of SKILL.md found no risky patterns, such as piping downloads into a shell, reading credential files or hidden Unicode. It is not a guarantee. Review the folder before installing.

What licence does Resume To Job Screen use?

Resume To Job Screen is published under the MIT licence (declared in SKILL.md). It allows redistribution, so the full SKILL.md is shown on this page.

How many tokens does Resume To Job Screen use?

About 1.5k tokens (SKILL.md is roughly 6k characters). Agents keep only the skill's name and description in context until a task matches; then they load SKILL.md in full.

What are the alternatives to Resume To Job Screen?

Skills that share tags, products or a category with Resume To Job Screen: Get Job (agentenatalie/get-job.skill, 632 stars), Resume Reviewer (weeelin98/ResumeDom, 173 stars), Build Resume Portfolio Site (tao943/build-resume-portfolio-site, 195 stars) and Cyber Resume Reviewer (mubix/cyber-resume-reviewer-skill, 184 stars). The comparison table on this page puts their stars, adoption, token cost, safety result and licence side by side.

Who maintains Resume To Job Screen?

mrmps (a GitHub user) maintains it in mrmps/classifier-dev, which has 424 GitHub stars. The repository holds 21 skills in this directory. The repository was last updated on October 7, 2026.

Source: mrmps/classifier-dev on GitHub. Facts on this page come from the repository at the commit we read; the author's words are quoted as theirs.