Help center
How can we help?
Product guides for admins, evaluators, and candidates — from first setup to the final decision file.
Getting started
Set up your workspace and run your first assessment.
Set up your workspace
Before you build your first assessment, take a few minutes to configure your organization and bring in your team. This guide covers the settings that affect every assessment you run.
Roles and permissions
SkillCort uses four workspace roles — owner, admin, recruiter, and evaluator — built on a least-privilege model. This article explains how access to reports, results, and the audit pack is scoped, and how every score is attributed.
Create your first assessment
This is the quick path from an empty draft to a published assessment: gather items, assemble the structure, attach rubrics, clear the publish checklist, and go live.
Invite your first candidates
Every assessment ships with a delivery — you configure it in the builder, publish, and invite candidates from the Deliver tab. This guide walks through delivery setup and what candidates see before they start.
Read your first results
Results in SkillCort flow from evaluator scores through weighted rollups to the Decision Board. This guide covers scoring basics, comparing candidates, drilling into evidence, and exporting reports.
Building assessments
Items, structure, rubrics, weights, and AI authoring.
Assessment builder overview
The assessment builder pairs a left sidebar with single-open accordion panels, so you work in one panel at a time while keeping the full section outline in view.
Sections, pages, and blocks
Every SkillCort assessment follows the same three-level structure: sections contain pages, and pages contain blocks. This article explains how to build and rearrange that structure, and how shuffle behaves at each level.
Question types overview
SkillCort supports twelve question types, from auto-scored multiple choice, matching, and ordering items to recorded video answers, code tasks, and rubric-only evaluations — plus AI-led interview and case-simulation formats. This overview groups them by what they capture and helps you pick the right type for each skill you want to measure.
Written and structured response questions
Written and structured response types let candidates produce an answer instead of selecting one. This article covers what each type captures and how these items are scored.
Media and technical question types
Media and technical question types collect richer evidence than text: recordings, photos, files, and code. This article explains what each type is for, what candidates experience, and what to expect at review time.
Multiple-choice questions and scoring
Multiple choice is SkillCort's most-used auto-scored question type, available as single-select or multi-select. This article covers answer keys, partial credit with per-option multipliers, negative marking, and CSV import for bulk authoring.
Rubrics: classic and matrix
Rubrics turn open responses into consistent, defensible scores. SkillCort supports two kinds — classic score-entry rubrics and level-based matrix rubrics — and this article explains how to build them, weight them, and understand what happens when you attach one to an assessment.
Scoring weights and role-fit
Role-fit is a weighted rollup, not a simple average. This article walks through the exact calculation and shows where you control each weight.
Generate items with AI
AI generation drafts items in bulk so you can build a library faster. It produces drafts, not finished questions — you review every item before it reaches a candidate.
Organize your item library
A tidy item library makes assessments faster to build and easier to audit. SkillCort organizes items with folders and folder-scoped tags, and a dedicated import modal brings them into your assessments.
Assessment copies vs. library items
When you add an item to an assessment, SkillCort stores a copy of it — not a live link to the library. This article explains what that means when you edit an item in the builder, edit it in the library, or want the two back in sync.
Edit a live assessment
You can keep editing an assessment after it goes live — add, edit, reorder, or remove items without pulling it offline. This article explains exactly what happens to candidates who already submitted, who is mid-attempt, and who starts after your change.
Delivering & security
Deliveries, invitations, time limits, and proctoring.
Set up a delivery
A delivery is how an assessment reaches candidates — and every assessment comes with one, created automatically the moment you create the assessment. You configure its schedule, access, and security in the builder, and manage candidates on the Deliver tab.
Proctoring and security settings
SkillCort's integrity controls are configured per delivery, disclosed to candidates before they start, and reviewed by people. This article lists every control and how to choose among them.
Lockdown mode and violation limits
Lockdown mode runs the attempt as a secure fullscreen exam in the browser — no extra software — and records when a candidate leaves it. A violation policy decides what happens: warn, pause, auto-submit at a limit, or flag for review.
What candidates see
This article walks through the candidate's side of a SkillCort assessment, from the consent gate to final submission. Use it to anticipate questions and set expectations before you invite anyone.
Invite candidates
Email invitations are the default way candidates reach an assessment. This article covers what has to be true before you can invite anyone, how the time window works, and what the candidate experiences after clicking the invitation.
Time limits: total and per page
SkillCort supports two kinds of time limits: a total assessment cap and per-page limits. This article explains how each is enforced and what candidates experience when time runs out.
Evaluating & deciding
Scoring, AI-assisted review, the Decision Board, and reports.
Evaluate responses
Evaluators score candidate responses against the shared rubric attached to the assessment. This article covers matrix and classic scoring, notes as evidence, and how scores from multiple evaluators combine.
AI-assisted review
AI assistance speeds up evaluation without taking it over: it summarizes long material and drafts score suggestions, and a named person always confirms or overrides. AI is decision support — never the decision-maker.
Use the Decision Board
The Decision Board puts candidates side by side on weighted role-fit and lets you drill from any score into the evidence behind it. It is where evaluation becomes a defensible decision.
Candidate evidence reports
A candidate evidence report is the single document that shows how a candidate performed and why the scores add up the way they do. Use it when you need the full evidence behind a role-fit percentage.
Export the audit pack
The audit pack is one self-contained decision file: everything a reviewer needs to reconstruct how a hiring decision was made, without access to the platform. Export it from the Reports page.
Security & data
Privacy, access control, integrity events, and AI provenance.
Data privacy and retention
SkillCort collects candidate data for one purpose — evaluating the assessment they were invited to — and gives each workspace control over how long that data lives. This article explains the principles behind collection, protection, and deletion.
Access control
SkillCort applies role-based, least-privilege access: each person sees only what their work requires. This article explains how roles scope access, how actions stay attributable, and how candidate-side access is controlled per delivery.
Integrity events explained
Integrity events are the signals SkillCort records during an attempt when a delivery has security settings enabled. They appear as a reviewable timeline in evaluation — context for a human decision, never an automatic verdict.
AI provenance and logging
Every AI output used in evaluation is logged with the model and version that produced it, and no AI suggestion becomes part of the record until a named person confirms or overrides it. AI is decision support — the decision stays human.
Still need a hand?
We help with setup, design, and rollout
Bring your questions to a live session — from your first assessment to a full hiring program.