integrations.sh
← all integrations

Resume Builder

MCP server openai

Resume enables ChatGPT to analyze, improve, and generate professional, ATS-friendly resumes. Upload a PDF or DOCX resume to identify key gaps, strengthen experience bullets, and refine impact with targeted suggestions. Resume then turns those improvements into a clean, structured, and downloadable PDF resume optimized for readability and applicant tracking systems — all within the ChatGPT conversation, without starting from scratch.

Homepage
https://jobright.ai
Remote URL
https://mcp.jobright.ai/mcp
Auth
NONE

Tools (3)

Extracted live via the executor SDK.

  • parser_resume

    Use this when the user uploads a resume file or asks to parse a specific uploaded resume. This is the ONLY way to parse and analyze resumes. Do NOT read or analyze the resume file yourself. After this tool returns, use the returned fileId when calling diagnose_resume or update_resume. If no resume file is available, ask the user to upload one before calling this tool.

  • diagnose_resume

    Use this when the user asks for feedback, scoring, or improvement suggestions for a resume that has already been parsed. Do NOT generate your own resume feedback or analysis — always use this tool instead. Call parser_resume first if there is no fileId yet, then pass the returned fileId.

  • update_resume

    Use this and MUST be called when the user asks to rewrite, improve, optimize, tailor, modify, update, regenerate, apply explicit edits, generate a revised version, rewrite specific sections, improve specific content, or create a revised version from concrete patch instructions for a parsed resume. Do NOT generate resume content yourself — always use this tool, THIS IS THE ONLY WAY, THIS IS IMPORTANT. Call parser_resume first and use the returned fileId as the "fileId" field. Build items from the parsed resume and the user's concrete edit instructions. Use this tool only when the user wants to modify resume content.

    Input must contain exactly two top-level fields: fileId and items. Params's name should be "fileId" and "items" respectively. items must be an array of patch operations defined by the schema. Do not include fileToProcess, parserResult, editableResume, targetRole, jobDescription, text, file, updates, or any other top-level field.

    Wrong top-level input examples: { parserResult: { ... }, fileId: "uuid-from-parser" } { file: "sediment://file_123", fileId: "uuid-from-parser", items: [...] } { text: { ... } } { fileId: "uuid-from-parser", jobDescription: "...", items: [...] } { fileId: "uuid-from-parser", updates:{} }

    Correct top-level input examples: { fileId: "uuid-from-parser", items: [...] } { fileId: "uuid-from-parser", items: [{ indexPath: "summary", action: "update", value: "Built and deployed ML systems that improved forecast accuracy by 18%." }] } { fileId: "uuid-from-parser", items: [{ indexPath: "skills.Machine Learning[3]", action: "add", value: "PyTorch" }] }