More screenshots

    Works with the AI Agents you already use

    Claude CodeClaude CodeCursorCursorCodex CLICodex CLIMuseMuseDotsDots+21 more

    ResumeProof Match

    1

    Evidence-backed resume–JD matching, truthful rewrites, and approval-gated delivery.

    $6.66

    Secure checkout via Stripe

    Money-back guarantee

    Converts to your local currency at checkout

    Security scanned
    ResumeProof Match

    ResumeProof Match

    Example session with this skill installed

    Compare my current resume with this Growth Marketing Manager JD. Show which must-have requirements are supported, need confirmation, or are missing; explain every conclusion with evidence from my resume; recommend only truthful edits; and stop for my approval before generating any PDF or DOCX.

    • Read your context and instructions
    • Compiled the resumeproof match

    # Recommendation: Apply after two checks The resume shows strong campaign ownership and measurable optimization work. The main uncertainty is not keyword coverage: it is whether partner activation and English collaboration can be supported by specific examples. I would verify those two points before finalizing the tailored resume. ## Three separate signals | Signal | Score | What it means | | --- | ---: | --- | | Role fit | 82/100 | Most core responsibilities are supported; one must-have needs confirmation. | | Evidence coverage | 76/100 | Several strong outcomes exist, but partner-management evidence is still thin. | | Document readiness | 68/100 | The facts are useful, but the current bullets do not make ownership and causality easy to scan. | ## Requirement-level evidence map | JD requirement | Status | Resume evidence | Decision | | --- | --- | --- | --- | | Own campaigns end to end | Supported | Managed channel brief, budget, launch schedule, and post-campaign review for paid media programs. | Keep and move higher. | | Optimize CPA and conversion funnels | Supported | Used audience and creative A/B tests to reduce registration cost by 28%. | Rewrite to connect action and result. | | Use analytics and tracking | Supported | Worked with GA4, UTM conventions, and weekly registration/deposit analysis. | Add the decision enabled by the analysis. | | Activate affiliates or partners | Needs confirmation | Resume names media channels but does not show partner recruitment, enablement, or retention. | Ask for a concrete example; do not imply ownership yet. | | Work comfortably in English | Needs confirmation | Cross-regional work is visible, but English as a working language is not explicitly evidenced. | Confirm before adding. | ## Proposed edit **Original** > Adjusted campaign mix through audience and creative A/B tests, reducing registration cost by 28%. **Suggested version** > Owned audience and creative A/B tests across paid campaigns, reallocating budget to the strongest combinations and reducing registration cost by 28%. **Why this change** The result was already credible, but the original did not show who made the decision or how the test changed allocation. The revision makes ownership, method, and outcome visible without adding a new claim. ## Do not add yet - Affiliate recruitment or retention ownership - English fluency as a confirmed qualification - Any metric that does not appear in the evidence ledger ## Next step Please confirm the two unresolved items above and review the complete text draft. I will not generate formatted files until you approve the full text.

    Connects securely to your tools. The creator never sees your data.

    About this skill

    ResumeProof Match is an end-to-end, evidence-first workflow for AI agents. It turns a resume and a target job description into an explainable application decision, a truthful tailored draft, and versioned final files that have been checked against the text the user actually approved. ## What it does - Classifies JD requirements as must-have, nice-to-have, or must-not. - Builds a claim-to-source evidence ledger from the resume and supporting material. - Maps every requirement to confirmed, needs-confirmation, or unsupported evidence. - Keeps role fit, evidence coverage, and document readiness as separate signals. - Produces a gap report, hard-gate recommendation, and evidence-backed edits. - Requires approval of the complete text before any formatted resume is generated. - Binds QA and delivery records to exact file hashes, preventing stale approvals or checks from being reused. ## Why it is different Keyword overlap alone cannot tell whether a candidate truly meets a requirement. ResumeProof Match preserves the original JD wording, points to the exact resume evidence behind each conclusion, and refuses to invent missing experience, tools, metrics, dates, titles, or language ability. Its score is an explainable rubric, not a fabricated ATS probability. ## How the workflow runs 1. Ingest a resume, JD, and optional supporting evidence. 2. Build and confirm the evidence ledger. 3. Review requirement-level matches, conflicts, gaps, and the apply / verify first / skip recommendation. 4. Review proposed edits with the original text, revised version, and reason. 5. Approve the complete plain-text resume. 6. Generate, inspect, QA, and deliver versioned PDF or editable files. ## Portable by design The deterministic helper uses the Python standard library for workspace setup, scoring, approval records, QA binding, and versioned delivery. Optional PDF text inspection can use pypdf or pdftotext. No account, database, vector store, or hosted backend is required for the skill itself. ## Use cases - Decide whether a role is worth applying for. - Find real qualification gaps before rewriting. - Tailor a resume without hallucinating achievements. - Maintain a review-first resume workflow across multiple applications. - Produce auditable final files that match the approved text. Live web app: https://resumeproof.szw19990924.chatgpt.site/match/new Source and documentation: https://github.com/caihhhhhh/resume-proof-match

    Requires a user-provided resume and job description, plus confirmation for any fact that is not already supported by evidence. It does not guarantee interviews or claim to predict an employer’s ATS. PDF text or visual inspection depends on the tools available in the host environment. The workflow prepares application materials but does not submit job applications on the user’s behalf.

    How to use

    1. 1

      Connect your agent to the Agensi MCP

      Copy the prompt and paste it into Claude, ChatGPT, Cursor or any agent that supports MCP. It sets up the connection for you. Manual setup

    2. 2

      Sign in to Agensi

      Your agent shows a sign-in link. Sign in and click Allow.

    3. 3

      Ask for what you need

      Your agent loads this skill and uses it on your task. It always asks before any payment.

    Reviews

    No reviews yet

    Be one of the first to try it. Every listed skill goes through the trust checks below.

    Security scanned

    Checked by our 8-point scan before listing

    Fresh listing

    Recently published to Agensi

    Money-back guarantee

    Doesn't do what its page promises? You get your money back.

    Trust & safety

    Security scan

    Security scan passed 1 month ago

    Requires

    Python
    Node.js

    Permissions

    Terminal / Shell
    Read Files
    Write Files

    Reads the user-provided resume, JD, and optional evidence files. Writes a local application workspace, approval and QA records, and versioned deliverables. No network access is required for the deterministic workflow.

    Listed1 month ago

    What's inside

    Frequently Asked Questions

    No. Every proposed claim is linked to confirmed evidence, marked as needing confirmation, or rejected as unsupported. Missing tools, metrics, dates, titles, and language ability are never fabricated.

    No. The score is an explainable decision aid built from JD requirements, evidence coverage, and hard gates. It does not pretend to know a company’s private ATS model or guarantee an interview.

    The gate lets the user review every word before formatting. Approval is recorded against the exact text hash, and later QA is bound to the exact generated files, so stale drafts or checks cannot be silently reused.

    Provide a resume, a target job description, and any optional evidence you want considered. Python 3.10 or newer is recommended for the deterministic helper; the core workflow uses the standard library.

    Yes. After the complete text is approved, the host agent can generate PDF or editable outputs, inspect the extracted text and page layout, record QA, and deliver versioned files. Available formats depend on the host environment.

    New in HR & Recruiting