Ransomware
HIGHEncrypted or deleted data stops planning, dispatch and billing.
Controls: Immutable backups, point-in-time restore, least-privilege access, DR runbooks.
For CISOs, CIOs, IT security and internal audit
Security for systems that run your plants, your money and your machines.
Manufacturing platforms carry formulations, costs, customer contracts and live machine connections. A breach is not just a data problem — it can stop a line. This page sets out the threats we design against, the controls we run and the evidence we can share with your security team.
AES-256 · TLS 1.2+
Encryption at rest and in transit
Per-tenant
Logical data isolation by default
Read-only
Default mode for machine connections
[Status]
Certification — see section 05
01 · What we defend against
A threat model written for manufacturing,
not borrowed from banking.
Formulation & cost leakage
HIGHRecipes, BOMs and margins are your competitive edge.
Controls: Field-level permissions, export controls, watermarking, access logs.
Compromised credentials
HIGHPhished or shared logins used by outsiders.
Controls: SSO with MFA, session limits, device sign-out, anomaly alerts.
Plant network exposure
HIGHA cloud link becomes a path into PLCs and SCADA.
Controls: Outbound-only edge gateways, read-only tags, OT/IT segmentation.
Insider misuse
MEDStaff or contractors accessing data beyond their role.
Controls: Role and plant scoping, maker-checker approvals, immutable audit trail.
Third-party & supply chain
MEDVulnerable libraries or vendor integrations.
Controls: Dependency scanning, signed builds, vendor security reviews.
Lost or stolen field devices
MEDPhones with farmer or customer data go missing.
Controls: Encrypted local storage, remote wipe of app data, PIN/biometric lock.
Tampered records
MEDWeights, lab results or approvals altered after the fact.
Controls: Versioned records, server timestamps, edit history with reason codes.
02 · Defence in depth
Nine layers of control. Pick one to see what’s inside.
Identity & access
Every person, device and integration is authenticated and limited to what it needs.
Single sign-on
SAML / OpenID Connect with your identity provider.
Multi-factor authentication
Enforced for admins and sensitive roles.
Role & scope permissions
By company, plant, warehouse, territory or module.
Maker-checker
Approvals for prices, recipes, payments and master changes.
Session controls
Timeouts, concurrent-session limits, forced sign-out.
Access reviews
Periodic review reports for your managers.
Evidence we can share: Role matrix, SSO configuration guide, access review report sample.
Application
The application enforces your rules on every request, not just on the screen.
Server-side authorisation
Every API call checked against role and scope.
Input validation
Protection against injection and cross-site scripting.
Field-level controls
Costs, margins and recipes hidden from roles that don’t need them.
Export controls
Downloads limited by role, logged and watermarked.
Audit trail
Who changed what, when, from where, and why.
Rate limiting
Abuse and brute-force attempts throttled.
Evidence we can share: Application security overview, sample audit log export.
Data protection
Your data is encrypted, isolated per tenant and recoverable.
Encryption at rest
AES-256 for databases, files and backups.
Encryption in transit
TLS 1.2+ for every connection.
Key management
Managed keys with rotation; customer keys on request.
Tenant isolation
Logical separation of every customer’s data.
Masking
Sensitive fields masked on screen and in exports.
Backups
Immutable, encrypted backups with tested restores.
Evidence we can share: Encryption summary, backup and restore test report.
Network
Only the traffic that must flow is allowed to flow.
Private networking
Databases and services never exposed to the internet.
Web application firewall
Common attacks filtered at the edge.
DDoS protection
Provider-level protection for public endpoints.
Segmentation
Separate zones for app, data and management.
Private connectivity
VPN or private links to your sites on request.
IP allow-listing
Admin access limited to approved networks.
Evidence we can share: Network architecture diagram, firewall policy summary.
Infrastructure
Hardened, patched and reproducible cloud infrastructure.
Infrastructure as code
Every environment built from reviewed templates.
Hardened images
Minimal, patched base images.
Patch management
Security patches applied on a defined schedule.
Environment separation
Production isolated from test and development.
Least-privilege ops
Engineers get time-bound, logged access.
Malware protection
Endpoint protection on managed systems.
Evidence we can share: Patch policy, environment architecture overview.
Plant & machine (OT)
Machines are connected without opening your plant network.
Outbound-only gateways
No inbound ports into the plant.
Read-only tags
Write-back only as a separate, approved change.
OT/IT segmentation
Gateways sit in a DMZ between levels.
Store and forward
No data loss during network outages.
Signed updates
Gateway software updates signed and scheduled.
Asset inventory
Every connected device and tag listed.
Evidence we can share: OT connectivity design, tag list template, gateway hardening guide.
Devices & field apps
Phones and tablets in the field carry as little risk as possible.
Encrypted local storage
Offline data encrypted on the device.
PIN / biometric lock
App-level lock on every launch.
Remote wipe
App data cleared when a device is lost.
Sync limits
Only the user’s own scope synced to the device.
Version control
Old, unsupported app versions blocked.
Device binding
Logins tied to registered devices on request.
Evidence we can share: Mobile security overview, device policy options.
AI safety
AI features see only what the user can see, and never train on your data for others.
Permission-aware AI
The assistant respects the user’s role and scope.
No cross-customer training
Your data does not train shared models.
Prompt & output logging
AI interactions recorded for review.
Human approval
AI suggests; people approve changes.
Data minimisation
Only the fields needed are sent to the model.
Provider controls
Model providers bound by contract and region.
Evidence we can share: AI data-flow description, model provider terms summary.
Monitoring & operations
We watch the platform around the clock and act on what we see.
Centralised logging
Application, access and infrastructure logs retained.
Alerting
Anomalies and failures raise alerts to on-call engineers.
SIEM export
Logs streamed to your SIEM on request.
Uptime monitoring
Health checks on every critical service.
Disaster recovery
Documented RPO / RTO with periodic DR tests.
Capacity planning
Headroom reviewed before peak seasons.
Evidence we can share: Monitoring overview, DR test summary, sample alert runbook.
03 · Plant & machine (OT) security
Connecting machines without opening your plant network.
Level 4–5 · Enterprise
Internet via TLS
ERP, Anvaya cloud, CXO dashboards
Business systems and cloud services
DMZ
Outbound only
Edge gateway
Buffers, converts protocols and forwards data outbound
Level 3 · Operations
Read from DMZ
Historians, SCADA servers
Site operations systems
Level 2 · Control
No cloud access
SCADA / HMI
Supervisory control
Level 0–1 · Process
No cloud access
PLCs, sensors, drives
Machines and instruments
No inbound ports
Nothing on the internet can initiate a connection into your plant.
Read-only first
We start with reading tags. Any write-back is a separate, approved change.
Plant engineering owns the change
Your controls team approves every tag, gateway and schedule.
Store and forward
Gateways buffer during outages so production data is not lost.
Signed updates
Gateway software updates are signed and scheduled.
Asset inventory
Every connected device and tag is listed and reviewed.
Aligned to IEC 62443 concepts
Zones and conduits approach for industrial networks.
04 · Who can see what
Access that follows your org chart
down to the plant, the ledger and the field.
Always Choose what level of access who needs
| Example role | Recipes & BOM | Costing & margin | Ledgers | Machine data | Farmer PII | Admin |
|---|---|---|---|---|---|---|
| CFO | View | Full | Full | View | — | — |
| Plant head | Full | View | — | Full | — | — |
| Production supervisor | View | — | — | View | — | — |
| Quality manager | View | — | — | View | — | — |
| Field officer | — | — | — | — | Own area | — |
| Store keeper | — | — | — | — | — | — |
| IT admin | — | — | — | — | — | Full |
Illustrative. Roles, scopes and approvals are configured to your organisation. Scope can be limited by company, plant, warehouse, territory or cost centre.
Joiner-mover-leaver
User provisioning and removal driven from your identity provider.
Break-glass access
Emergency access logged and reviewed.
Trikaan staff access
Support access only with your approval, time-limited and logged.
Service accounts
Integration credentials scoped, rotated and monitored.
Segregation of duties
Conflicting permissions flagged at assignment.
Access review export
Quarterly reports for managers to certify.
05 · Framework alignment
Mapped to the frameworks your auditors use.
Our controls are organised against widely used frameworks so your team can map them to your own policies. Alignment is not certification — current certification status is shown below.
| Control area | What we do | ISO/IEC 27001 theme | NIST CSF function | OT reference |
|---|---|---|---|---|
| Access control | SSO, MFA, role and scope permissions, access reviews | Access control, identity | Protect | Zones & conduits |
| Cryptography | Encryption at rest and in transit, key management | Cryptography | Protect | — |
| Operations security | Logging, monitoring, patching, malware protection | Operations security | Detect | Monitoring |
| Network security | Segmentation, WAF, private connectivity | Network security | Protect | Segmentation |
| Secure development | Code review, scanning, security testing | Secure development | Protect | Secure lifecycle |
| Supplier relationships | Sub-processor reviews, contracts | Supplier security | Identify | — |
| Incident management | Runbooks, notification, post-incident review | Incident management | Respond | Incident response |
| Business continuity | Backups, DR, tested restores | Continuity | Recover | Recovery |
| Asset management | Asset and tag inventory, data classification | Asset management | Identify | Asset inventory |
06 · Secure development
Security checks in every build, not once a year.
- 1
Threat modelling
For new modules and integrations.
- 2
Secure coding standards
Guidelines and training for engineers.
- 3
Peer code review
Every change reviewed before merge.
- 4
Static analysis
Automated code scanning on every build.
- 5
Dependency scanning
Known-vulnerable libraries blocked.
- 6
Secrets detection
Builds fail if credentials are found.
- 7
Dynamic testing
Automated scans against staging.
- 8
Release approval
Change approval before production.
07 · Vulnerability management
Found, ranked and fixed to a clock.
| Severity | Remediation target |
|---|---|
| Critical | 1 days |
| High | 2 days |
| Medium | 5 days |
| Low | Next planned release |
[Targets are placeholders; confirm against internal policy.]
Independent penetration tests
[Frequency] by an external firm; summary shared under NDA.
Customer-led testing
Permitted with notice and agreed scope.
Responsible disclosure
Researchers can report issues safely.
Remediation tracking
Every finding tracked to closure.
08 · Shared responsibility
Clear lines between what we own and what you own.
| Area | Trikaan cloud | Your cloud account | On-premise |
|---|---|---|---|
| Physical data centre | Provider | Provider | You |
| Cloud infrastructure | Trikaan | You | You |
| Platform patching | Trikaan | Trikaan | Shared |
| Application security | Trikaan | Trikaan | Trikaan |
| Backups & DR | Trikaan | Shared | Shared |
| Identity provider & users | You | You | You |
| Role assignment | You | You | You |
| Plant network & OT | You | You | You |
| Edge gateway | Trikaan | Shared | Shared |
09 · Incident response
If something goes wrong, you hear it from us first — with facts.
Detect
Alert raised
Automated monitoring or a report triggers triage.
Within [X] hours
Contain
Isolate affected systems, preserve evidence.
Within [X] hours
Notify you
Named contact informed with known facts and impact.
Ongoing
Investigate
Root cause, scope and affected data established.
As agreed
Recover
Restore service and data; verify integrity.
Within [X] days
Report
Written post-incident report with corrective actions.
Regulatory timelines. Indian rules require certain cyber incidents to be reported to CERT-In within short timeframes, and personal data breaches to be notified under the DPDP framework. Our incident process is designed to give you the facts you need to meet your own obligations, and we meet ours as a service provider. [Confirm exact notification commitments with legal before publishing.]
Security review support
Your security review, made shorter.
Send us your questionnaire. We return it with evidence and a call with our security lead, typically within [X] business days.
Start a security review →Found a vulnerability? Report it to [security@ email]. We acknowledge responsible disclosures within [X] business days.
Security whitepaper
Architecture and controls overview
Filled questionnaire
Standard security questionnaire answers
Pen test summary
Latest independent test, under NDA
Data flow diagram
What data goes where
Sub-processor list
Vendors and their roles
Incident response plan
Roles, timelines, notification
DR & backup summary
RPO, RTO and test results
Policies index
Information security policy set
Questions security teams ask
Can our data be hosted in India?
Yes. Indian customers are hosted in India cloud regions by default. Other regions, your own cloud account or on-premise deployment are available on request.
Can we use our own encryption keys?
Customer-managed keys are available on request for supported deployments, so you can control and rotate the keys that protect your data.
Can Trikaan staff see our data?
Only with your approval. Support access is time-limited, scoped to the issue and logged, and you can review those logs.
How do you protect our plant network when connecting machines?
Edge gateways make outbound-only connections from a DMZ, read tags by default and never open inbound ports. Your plant engineering team approves every tag and change.
Can we run our own penetration test?
Yes. Customer-led testing is permitted with notice and an agreed scope, and we track every finding to closure.
Do you support our SIEM?
Yes. Application, access and audit logs can be streamed or exported to your SIEM in standard formats.
What happens to our data when the contract ends?
You get a full export in open formats, after which your data is deleted from production and expires from backups on schedule, with a deletion certificate.
Are you certified?
Our controls are mapped to ISO/IEC 27001, NIST CSF and IEC 62443 concepts. Current certification status and audit evidence are shared in the security pack under NDA.
