Overview

Real-time AI usage, cost and security monitoring

// CONTROL STATUS ACTIVE
ESTIMATED SAVED TODAY
$—
estimated · process lifetime
Tokens Observed
—
AutoFix Applied
—
Risk Blocked
—
// SMART ROUTING OBSERVING
Standard
—
Low Cost
—
Large Context
—
High Risk
—
Enforcement Activations
—
Overrides Observed
—
Cheap-Tier Activity
—
Conservative Activity
—
Recent Routing Decisions
    // TOKEN CONTROL PIPELINE OBSERVING
    Requests Analyzed
    —
    Est. Tokens Saved
    —
    Avg. Optimization
    —
    Avg. Request Size
    —
    Top Waste
      Token Budget Activity
      Near Budget
      —
      Over Budget
      —
      Caps Applied
      —
      Rejects
      —
      Tokens Saved
      —
      All-time
      Cost Prevented
      —
      All-time
      Risk Blocked
      —
      All-time
      AutoFix Applied
      —
      All-time
      // LIVE CONTROL FEED
      • Loading governance activity…
      Cache Hits
      —
      Requests served from cache
      Tokens Saved
      —
      Via prompt cache
      Cache Entries
      —
      Unique prompts cached

      Cost by Department

      Requests by Model

      Daily Usage — Last 30 Days

      Audit Logs

      All AI requests passing through Nizam Control

      TimeUserModelDept TokensCostRiskMaskedCache

      Security Alerts

      Requests with sensitive data detected

      TimeUserModelRisk ScoreMasked FieldsAlert

      Cost Intelligence

      AI spending breakdown by department and model

      Department Breakdown

      Model Breakdown

      Department Summary

      DepartmentRequestsTokensCost (USD)

      Budget Management

      Set monthly spending limits per department

      💡 Budget limits are enforced in real-time. Requests that would exceed the monthly limit are automatically blocked with a 402 error.

      Set Department Budget

      Active Budgets — This Month

      API Key Management

      Create and manage user API keys

      Create New Key

      All API Keys

      NameRoleDepartmentAPI KeyRate LimitStatusActions

      Users

      Manage dashboard users and access

      Add New User

      All Users

      NameEmailRoleDepartmentCreatedAction

      Playground

      Test AI models through Nizam Control in real-time

      Configuration
      Prompt
      Security Analysis
      Run a prompt to see security analysis...
      Usage & Cost
      —
      Model Routing
      —
      Response
      Run a prompt to see the response...

      Architecture

      How Nizam Control sits between your apps and AI providers

      Request Flow

      💻
      Your Application
      Developer / App / Service
      ↓
      🛡️
      Nizam Control Gateway
      Auth · Rate Limit · Policy · Token Limit
      ↓
      🧠
      Intelligence Layer
      Budget Check · Model Downgrading · Cache
      ↓
      🔒
      Security Layer
      PII Masking · Prompt Firewall · Risk Scoring
      ↓
      🤖
      AI Providers
      OpenAI · Gemini · Anthropic Claude
      ↑ Response + Audit Log
      🔐
      Identity & Access
      API key auth with role-based model policies. Intern → basic models, Admin → all models.
      💰
      Budget Enforcement
      Monthly per-department budgets. Requests that would exceed limits are blocked before reaching the LLM.
      ⚡
      Model Downgrading
      Simple prompts auto-routed to cheaper models. Save up to 90% on basic queries without user awareness.
      🗄️
      Prompt Cache
      Identical prompts served from cache. Zero token cost on repeated queries.
      🔒
      Security Scanning
      PII auto-detected and masked. Prompt injection and jailbreak attempts blocked.

      Integration — 3 Lines of Code

      // Before Nizam Control
      fetch('https://api.openai.com/v1/chat/completions', { ... })

      // After Nizam Control — governance, audit, AutoFix, economics included
      fetch('<your-edge-url>/v1/chat/completions', {
        headers: { 'authorization': 'Bearer sk-...' },
        body: JSON.stringify({ model: 'gpt-4o-mini', messages: [...] })
      })

      Governance Settings

      Durable per-workspace defaults. Conservative by default. Edge enforcement bridge ships in Task 21.

      // WORKSPACE POLICY PERSISTED
      AutoFix Default Mode
      Observe by default. Enforce only when explicitly enabled.
      Routing Enforcement
      When off, smart routing only observes. When on, opt-in headers may shape execution.
      Cheap-Tier Mode
      Cheap-tier mode reduces output budget for low-risk requests.
      Conservative Mode
      Conservative mode avoids aggressive mutation for risky requests.
      Force Observe-Only
      Outranks everything. No enforcement, no mutation. Recommended during pilot evaluation.
      PII Detection
      PII detection flags sensitive patterns before upstream execution.
      Max Output Tokens
      Optional cap applied when cheap-tier enforcement triggers. Leave blank for no cap (16–8192).
      Token Budget Limit
      Soft monthly budget hint. Surfaces in the economics ledger. (128–250000)
      V1 honesty: these settings are persisted durably and shown here. The Edge proxy does not yet consume workspace policy — request-time enforcement still uses explicit headers (x-nizam-autofix-mode, x-nizam-routing-enforce, etc.). The bridge ships in Task 21.

      Providers

      Bring Your Own API Keys — your keys, your data

      BYOK — Bring Your Own Key

      Add your own OpenAI, Anthropic, or Gemini API keys. Nizam Control will use your keys instead of the shared pool — your data never touches our credentials.

      Add Provider Key

      Configured Providers

      No providers configured — using Nizam Control shared keys

      Super Admin

      All companies and global platform stats

      Total Companies
      —
      Total Requests
      —
      Total Revenue Cost
      —

      Companies

      Company Email Plan Keys Requests Cost Joined