Using in workflows

Read typed Objects, traverse links, and dispatch actions from a workflow — the same blocks any other agent uses.

Workflows use the ontology through three tools, all exposed by the scrydon:ontology product. They show up in the Tools picker of any Agent block.

The three tools

Tool	Purpose
`Get Ontology Object`	Fetch a single typed Object by id.
`Traverse Ontology Links`	Given an instance and a link type, return the connected instances.
`Run Ontology Action`	Dispatch a typed action against an instance.

Each tool is typed — the LLM sees the object types and link types your organisation has defined, and it can call them by name.

Get Ontology Object

Input:
  type:     "RegulatedEntity"
  id:       "e1a2b3c4-0001"
Output:
  instance: { id, legalName, riskClassification, headquartersCountry, … }
  provenance: { bindingId, materializedAt, dlpLabels }

Properties the caller doesn't have clearance to see are masked according to the column's mask strategy (redact, null, deny). The output also carries the source-row provenance so the agent can cite where the answer came from.

Traverse Ontology Links

Input:
  fromType: "RegulatedEntity"
  fromId:   "e1a2b3c4-0001"
  linkType: "supervisedBy"
  page:     1
  pageSize: 50
Output:
  instances: [ { id, … } ]
  nextPage:  2  // or null when exhausted

Each connected instance is a fully typed Object (same projection as Get Object). Masking and row filters apply.

Run Ontology Action

Input:
  type:    "AssignAsset"
  input:   { assetId, userId }
  idempotencyKey?: "..."
Output:
  result:  { assetId, ownerId, assignedAt }
  // or error: { kind: "PreconditionFailed", rule: "asset-unassigned", … }

Action dispatch passes through the policy decision point. Failures come back as typed errors the agent can branch on.

The Agent block

In an Agent block, attaching the scrydon:ontology tools gives the agent access to all three. The LLM sees the object types and link types as part of the tool schema; it can plan calls like "find the regulator that supervises Acme Holdings" without hardcoded table names.

Common patterns

Retrieval-grounded answer

User: "What's the risk classification of Neptune Capital?"
Agent → Get Ontology Object(type=RegulatedEntity, lookup=Neptune Capital)
Agent → returns risk classification + provenance citation

Multi-hop traversal

User: "Which regulators supervise high-risk entities in Belgium?"
Agent → Search RegulatedEntity (filter: country=Belgium, risk=high)
Agent → For each: Traverse "supervisedBy" link
Agent → Aggregate distinct regulators

Action with guardrails

User: "Assign asset A-123 to Maya"
Agent → Run Ontology Action(AssignAsset, { assetId: A-123, userId: Maya })
Action → PreconditionFailed: asset-unassigned (asset is currently assigned to John)
Agent → reports the failure, asks the user to confirm reassignment

What the agent doesn't see

Raw column names. The agent operates on typed properties (riskClassification), not on risk_class columns.
Other tenants' data. Authorisation is enforced at every tool call.
Masked values. Properties the calling user can't see come back masked.
Actions the calling user can't dispatch. The tool schema is filtered to the actions the caller has grants for.

Object types — what the agent reads.
Link types — what traversal walks.
Action types — what dispatch executes.
Blocks → Agent — where the tools are attached.

Using in workflows

On this page