Skip to content
7lens Docs
7lens.io Register to Test

Architecture

7lens reads what the tools you already use — CMDB, cloud, security, monitoring, finance, application catalogs, custom systems — already know, and connects it inside your own environment. It does not replace those tools, and it does not copy your data out.

Select a part of the diagram to see what it does, or press Follow a fact to walk one fact from a source system to the people and agents who read it.

The 7lens architecture. Your existing systems (cloud, security, finance, CMDB, operations and other sources) send what they know to Collector. Everything else runs in your environment, with zero data egress: Collector records and checks each delivery and passes it to the Semantic Engine, which uses the Ontology to normalize, reconcile, validate and map it, keeps the trail and resolves disagreements, and produces your enterprise Context. A 7lens management plane (SaaS), outside your environment, receives human configuration only and no customer data. AI proposes; trusted data, fixed rules and an explicit check decide what is true. People read the Context through Lenses 1 to 6, AI agents through Lens 7 over MCP and the API, and software through the API.

Your existing systems

7lens reads what they know. It does not replace them.

  • CloudCloud platforms
  • SecuritySecurity tools
  • FinanceFinance systems
  • CMDBAsset and configuration records
  • OperationsMonitoring tools
  • Other sourcesApplication catalogs · Custom tools
7lens SaaS · outside your environment

Management plane

Human configuration only. No customer data.

Runs in your environment · zero data egress

Collector

The front door for what your systems already know.

  • Records who sent it
  • Checks the payload is usable
  • Accepts a full load or a change
  • Keeps a status for each delivery

Ontology

The model: what kinds of things exist and how they can connect.

  • Organization
  • Groups
  • Technology
  • Business
  • Extensions: Inventory, Contracts, Processes

Semantic Engine

Turns source information into one connected picture.

  1. Normalize
  2. Reconcile
  3. Validate
  4. Map
  5. Keep the trail
  6. Resolve disagreements

Your enterprise Context

The model filled in for your enterprise.

  • Organization
  • Groups
  • Technology
  • Business
  • Inventory
  • ContractsFuture
  • ProcessesFuture

AI proposes. Trusted data, fixed rules and an explicit check decide what is true.

Seven Lenses. One Context.

Each Lens asks its own question of the same picture.

People

Lenses 1 to 6, written for the people who lead and govern.

  • 01Security Lens
  • 02CTO Governance Lens
  • 03CFO Capital Lens
  • 04CIO Strategy Lens
  • 05Lock-In Free Lens
  • 06Automation & AIOps ROI Lens

AI agents

Lens 7, through MCP and the API, under the same permissions and audit trail as a person.

  • 07AI Lens

Software & systems

The same Context through the API: what exists, how it connects, and where each fact came from.

  • Custom Lens
  • Custom Lens
  • Custom Lens

Follow a fact: Example fact (illustrative)

  1. Your tools keep their facts and stay the authority for them. Here, a cloud platform and a CMDB each describe the same cluster. (Cloud platform: prod-eu-k8s-01; CMDB: Production K8s (EU))
  2. Collector records who sent each delivery, checks the payload is usable, accepts a full load or a change, passes it to the Semantic Engine and keeps a status. (2 deliveries received; Sender recorded; Payload checked; Status kept)
  3. The Semantic Engine does not treat incoming data as already true. First it normalizes names and formats into the 7lens model. (Names and formats aligned to the model)
  4. Reconcile: records that describe the same real thing become one canonical record. (2 records → 1 canonical record)
  5. Validate and map: rules mark anything that does not fit the model, and each fact is attached to the right place. AI may suggest the mapping; trusted data, fixed rules and an explicit check decide. (Checked against the Ontology; Asset Type: Kubernetes cluster; Belongs to a Platform; Suggested by AI, decided by rule and check)
  6. Keep the trail and resolve disagreements: the record keeps where each fact came from, when it was seen and how confident the match is. If two sources conflict, a defined rule picks the outcome. Nothing is chosen quietly. (Sources kept; When it was seen; Match confidence; Conflicts decided by a rule)
  7. The result is one connected fact in your enterprise Context, built and kept inside your environment. The source systems are unchanged. (One connected Asset; Source systems unchanged)
  8. The same Context is read through the seven Lenses: Lenses 1 to 6 by people, Lens 7 by AI agents through MCP and the API. Software reads the same Context through the API. (Lenses 1–6: people; Lens 7: AI agents (MCP, API); Software: API)

Where it runs

Section titled “Where it runs”
  • In your environment. The data, the filled-in model, the processing, and the Context stay there. That includes sign-in with your identity system, permissions by role, separate permissions for people, AI assistants, and automations, a full audit trail, and data kept in the required region.
  • Customer data means the Context and everything used to build it: what sources sent, how it was mapped, the reconciled records, and how the lenses are configured.
  • Management plane (SaaS). It handles only the part of the core semantic concepts that 7lens needs to operate, driven by human configuration. No customer data reaches it.

Authority over facts

Section titled “Authority over facts”

Go deeper

Section titled “Go deeper”
  • Product definition: how information gets in, how a source fact becomes part of the picture, and how people and software use it.
  • Data model: the Ontology that the Semantic Engine applies.