Resume Deep Analysis and Interview Assessment Report

Problem

When screening resumes, the hard part is not extracting names, schools, or project keywords. It is deciding whether an experience can be verified: whether full-time roles are all under 6 months, whether gaps exceed 2 months, whether claimed automation testing has real output, and whether the candidate led implementation or only executed scripts. Rule-based keyword matching can miss context and mistake stacked buzzwords for competence.

How It Works

The skill works on a single resume PDF. It first extracts the full text, then analyzes fixed dimensions. Basic information is broken down into name, school, major, degree, graduation date, current company, total experience, language ability, and certifications. Each work entry lists company, title, period, and exact duration in months, with checks for stability, gaps, and the boundary between pre-graduation internships and post-graduation full-time roles.

For strengths and weaknesses, it checks patterns such as certifications, domain experience, development stacks, hardware and testing tools, quantified metrics, academic signals, and stability, requiring several sentences of explanation for each item. The interview section produces 4-5 questions covering the biggest doubt, core strengths, a live coding or diagram task, and SQL or Linux basics, with the purpose of each question stated. Finally, it writes an HTML report containing the verdict, timeline, strengths, weaknesses, interview questions, focus points, and a one-line summary.

Limits

The skill expects the resume PDF to be in the configured directory and for the user to provide the candidate name or file path. It is useful as a pre-interview review aid, not a replacement for background checks, compensation evaluation, or team fit assessment. Because the analysis relies on semantic judgment, output quality depends on resume completeness; if a full result already exists for the same candidate, reuse it instead of re-extracting.