Skills Matrix & Competency Tracker: Finally Know Who Can Do What
Build a governed skills matrix where employees self-assess their competencies, managers validate each rating before it counts, and the team's coverage and gaps light up against role requirements — with single-points-of-failure flagged and gaps linked to development actions.
A logged-in tool where you define a competency framework and role requirements, employees self-assess their skills, managers validate each rating before it becomes official, and a live matrix shows team coverage, gaps against role requirements, and single-points-of-failure — with gaps linked to development actions, every change audited, and a clean CSV export of the skills matrix.
Before you start
- A Supabase account (free)
- A Vercel account (free)
- A Resend account (free)
- Your competency framework (skills + proficiency levels) and role-to-skill requirements, even rough
- A roster of employees, their roles, and managers (CSV or a Google Sheet is fine)
- Claude Code or any AI coding agent
The problem this kills
"Who can run the month-end close if Priya's out?" "Do we have enough people certified on the new line before we scale it?" "Where are we one resignation away from a hole?" These are the questions L&D leaders and managers get asked in the worst moments — a sudden departure, a new project, an audit — and almost nobody can answer them quickly.
The usual answer is a spreadsheet skills matrix: names down the side, skills across the top, a smear of 1-to-5 ratings in the middle. It rots fast. The ratings are whatever people claimed about themselves two years ago, with no one to check them. Nobody knows which numbers are self-reported guesses and which a manager actually confirmed. Critical skills held by exactly one person hide in plain sight. And the matrix never connects to anything — you spot a gap, and there's no path from "we're thin on this" to "here's the training that fixes it."
A skills matrix is only useful if you can trust the numbers and act on them. That means a clear line between what someone claims and what a manager has validated, a coverage view that surfaces gaps and single-points-of-failure automatically, and a link from every gap to a development action. That's a real, governed application — not a colour-coded workbook.
What you'll build
A simple internal web app for L&D and managers. First you define your competency framework — the skills that matter and what each proficiency level means (e.g. 0 = none, 1 = aware, 2 = working, 3 = proficient, 4 = expert). Then you map role requirements: the role of "Senior Underwriter" requires level 3 in these five skills, level 2 in those three.
Employees self-assess their proficiency against the framework. Their manager then validates each rating — confirming it, raising it, or knocking it down — and only the validated level is recorded as official. The tool keeps self-assessed and validated levels side by side so you always know which is which.
From there it renders the matrix: team coverage by skill, gaps where people fall short of their role's required level, and a flag for any single-point-of-failure — a critical skill only one person holds. Each gap can be linked to a development action (a course, a mentor, an on-the-job assignment) so the matrix drives a plan, not just a heatmap. And the whole thing exports to a clean CSV.
What's inside the Implementation Plan
The downloadable plan is a step-by-step file you paste into an AI coding agent. It opens by interviewing you about your business — how you define skills and levels today, what your roles are and which skills they actually require, who validates whom, which skills are business-critical, your headcount and team structure, and the messy exceptions (people in two roles, contractors, expiring certifications, skills that decay). It reflects a short tailored spec back to you and gets your thumbs-up before it builds anything, so the matrix matches your competencies and your roles — not a generic template.
From there it walks the agent through the framework definition, role requirements, the self-assessment flow, the manager validation gate, the coverage-and-gap matrix, single-point-of-failure detection, the link from gaps to development actions, and the CSV export. Every step ends with a ready-to-copy prompt. There's a full "No API yet?" path that imports assessments from a Google Sheet / CSV and exports a clean matrix CSV — so you can build and run the whole thing this weekend regardless of what HRIS you're on.
The governance it includes (this is the point)
This is people data, so the controls aren't optional. The plan builds in login so only your team can use the tool, row-level security so managers and employees only ever see their own organization's data, a complete audit trail of who assessed what and who validated it and when, a hard human-validation gate so a self-assessed rating never becomes official until a manager signs off (and changes to the framework itself are approved by L&D), and duplicate guards so each employee has exactly one validated rating per competency.
Who it's for
L&D leaders, capability and training managers, team leads, and HR business partners — anyone who needs a trustworthy answer to "who can do what, and where are we exposed?" and is tired of a stale spreadsheet nobody believes. If you can describe your skills, your roles, and who validates whom, you can build this.
You've got this — start with the plan, paste the first prompt, and answer the interview. You'll have your first validated matrix on screen before the weekend's out.