Skip to main content

Evalium Core features — Mindmap (reference)

This is a compact, hierarchical view of Evalium’s core features and how they map to the platform’s execution-ledger model.

    1. Content authoring
    • 1.1) Question Management
      • CRUD + versioning (immutable once published)
      • Flexible content model (MCQ, MRQ, etc.)
      • Tagging (skills, topics, difficulty)
    • 1.2) Evaluation authoring (Assessment Blueprint)
      • Structure: Evaluation → Version → Section → Item
      • Content types
        • Fixed items (specific questions)
        • Buckets (random selection by tags)
        • Passages (reading text + linked questions)
      • KOE flavors
        • K (Knowledge): standard tests and exams
        • O (Observation): assessor-led practical checklists (partially implemented)
        • E (Evidence): artefact submission and review (implemented)
    1. Delivery & Execution
    • 2.1) Assignments
      • Targeting: users, emails, groups
      • Configuration: time limits, attempts, overrides
      • Security: lockdown browser requirements
    • 2.2) Delivery Sessions (Live Execution)
      • Lifecycle: start, progress tracking, auto-expiry/submit
      • Telemetry: focus/lockdown signals and session events
      • Resilience: auto-submit expired sessions
    • 2.3) Verification & Trust Levels
      • L1–L4 verification policy
      • Step-up context enforcement when required
      • Proctor gate for high-rigor sessions
    1. Results & Submissions (Execution Ledger)
    • 3.1) Submission Processing
      • Immutable records (amend, don’t edit)
      • Version snapshots (frozen evaluation structure)
      • Auto-scoring + pass/fail outcomes
    • 3.2) Ledger Events (Append-Only)
      • Evidence decisions, amendments, voids
      • Verification and ratification events
    • 3.3) Evidence Handling
      • Standalone evidence (zero-item submissions)
      • Inline evidence (item-linked)
      • Mixed evidence (answer + artefact)
      • Review lifecycle: approve/reject/void
    • 3.4) Results Remediation
      • Append-only corrections (new score versions)
      • Rules: mark_correct, drop_item, replace_key
    • 3.5) State Hashing
      • Submission/engagement state hashes for integrity
    1. Programmes & Engagements
    • 4.1) Programmes
      • Multi-step journeys (K/O/E bundling)
      • Progress tracking + completion status
      • Automated certification on completion
    • 4.2) Engagements (partially implemented)
      • Project-level tracking
      • Timeline over assignments + submissions
    1. Reporting & Analytics
    • Detailed reports
      • Session attempt report (single submission deep dive)
      • Evaluation summary (pass rates, distributions)
      • Question health (item analytics)
    • Data access
      • CSV exports
      • Glass Box API (read-only, ledger-backed)
    • Reporting projections
      • Projection schema + worker pipeline
      • Read model for dashboards/BI
    1. User & Access Management
    • RBAC (roles + capabilities)
    • Multi-tenancy + org isolation (RLS)
    • Authentication (HttpOnly sessions, magic links)
    • Link Principals (Glass Box access)
    1. Compliance & Security
    • Compliance Centre (DSAR, forget user, retention)
    • Audit logs (auth/admin/ledger actions)
    • Evidence storage tier lifecycle (HOT/ARCHIVED/DELETED)
    • Privacy receipts + artifact hashes in the compliance ledger