Skip to main content

Documentation Index

Fetch the complete documentation index at: https://mintlify.com/MadsLorentzen/ai-job-search/llms.txt

Use this file to discover all available pages before exploring further.

The /upskill command compares your candidate profile against the job postings you’ve been tracking and produces two outputs: a prioritized skill gap heatmap that shows which missing skills appear most frequently across your target roles, and a structured learning plan with concrete study resources, study directions tailored to your existing background, and realistic time-to-proficiency estimates. Resources are found via live web searches so they reflect current course offerings rather than stale recommendations. Use /upskill between application cycles for career planning, or immediately before a specific application to prepare for a role you know you want.

Usage

/upskill
/upskill https://jobindex.dk/job/1234567
Invoking /upskill without arguments runs in aggregate mode over all jobs in your tracker. Passing a URL runs in targeted mode against that single posting.

What You Get

Regardless of mode, /upskill produces a structured report saved to your upskill/ folder: Gap Heatmap A prioritized table of every skill gap identified, labeled by priority (Critical / High / Medium / Low), skill type (Hard / Domain / Tooling / Soft / Credential), and the source of the gap — for example 4/5 jobs, score 3.2 in aggregate mode, or required / preferred in targeted mode. The heatmap is printed to the terminal as an intermediate output before the learning plan begins. Learning Plan For every Critical and High gap (and Medium gaps when the total gap count is small), the learning plan provides:
  • 2–3 study resources found via live web search — courses, documentation, or books — with a one-line rationale for each
  • A study direction tailored to your background, specifying what to skip and where to start given what you already know
  • A time-to-proficiency estimate (e.g. ~20h, ~40h for a solid foundation)
Resources are grouped by theme (Cloud & Infrastructure, MLOps, Domain Knowledge, etc.) rather than listed alphabetically. Every web search includes the current year to avoid stale results. Suggested Study Order A numbered sequence that respects dependencies (prerequisites come first), prioritizes Critical over High over Medium, and places domain knowledge last so it can be studied alongside practical projects.

Broad vs. Targeted Analysis

/upskill operates in two distinct modes depending on how you invoke it: Aggregate mode (/upskill with no argument) analyzes all jobs in job_search_tracker.csv. It builds a frequency-weighted skill gap map — skills that appear in more jobs, and in jobs where your fit rating was lower, score higher. This mode is suited to career planning: understanding what to build over the next few months to become more competitive across your target market. Targeted mode (/upskill <URL>) fetches a single job posting and analyzes only that role’s required and preferred skills. Required skill gaps are flagged as Critical or High; preferred gaps as Medium. This mode is suited to focused interview prep: knowing exactly what to brush up on before applying to or interviewing for a specific role.

Output Location

Every report is saved as a Markdown file in the upskill/ folder:
  • Aggregate reports: upskill/report-YYYY-MM-DD.md
  • Targeted reports: upskill/report-YYYY-MM-DD-<company-slug>-<role-slug>.md (e.g. upskill/report-2026-04-20-guardsix-senior-ai-engineer.md)
When you run aggregate mode again later, /upskill loads the most recent previous report and adds a Since Last Report diff section showing which gaps you’ve closed (skills added to your profile) and which new gaps have appeared from jobs tracked since then.
Use /upskill for career planning between applications, not just immediately before applying. Running it once a month gives you a rolling diff of where your profile is strengthening relative to your target market — and makes the learning plan easier to act on gradually rather than cramming before a deadline.

Build docs developers (and LLMs) love