Oscar Research · Capabilities, Automation & Integration

A growing catalogue of callable public sector data and AI capabilities

Oscar can retrieve, validate, enrich, classify, compare, summarise and transform public sector information using a combination of Oscar data, open sources and AI.

The capabilities below are examples we have already implemented. They are not fixed products or the limit of what Oscar can do. New capabilities can be configured around a client's specific information requirement, workflow or use case.

What is an Oscar capability?

An Oscar capability is a reusable data or AI function built around public sector information. It may retrieve information from Oscar datasets or open sources, apply validation, matching, classification, summarisation or other processing, and return a useful result that can be consumed by a person, system, application or AI agent.

01 · REQUIREMENT

Start with the client's need

What information needs to be found, checked, enriched, compared, monitored, classified, summarised or transformed?

02 · SOURCES

Use the right public sector evidence

Draw from Oscar's researched datasets, public/open sources or a combination of both depending on the requirement.

03 · INTELLIGENCE

Add logic or AI where useful

Apply matching, validation, extraction, classification, summarisation, scoring or other processing to turn source information into a usable result.

04 · ACCESS

Expose the capability where it is needed

Make it available through an API, integration, bespoke route, PSN, feed, application or emerging MCP-enabled access model.

The catalogue is evidence of what has been implemented — not a closed menu.

The capabilities listed below are examples Oscar has already been able to deliver. A new client requirement may produce a new capability, new inputs, new outputs or a new access route.

Capability catalogue

Explore what Oscar can already retrieve, process and expose

Use the four views below to understand the same underlying capabilities from different angles: what they do, how they can be consumed, how they can fit into systems, and the technical detail available to developers.

Implemented examples

Capabilities

Reusable public sector data and AI functions Oscar has already implemented. These examples show the kinds of retrieval, validation, enrichment and intelligence tasks the underlying framework can support.

CapabilityWhat it doesBusiness problemAvailable throughMore
Loading capabilities...

Access & licensing

How a capability can be consumed

The same underlying capability may be exposed through different routes. This view separates the function itself from the access, commercial and delivery model.

CapabilityPSNAPIConnectorMCPBespokeLicensingMore
Loading access options...

Integrations

Use a capability inside another system or workflow

Where a capability needs to sit inside a CRM, product, application, data pipeline or operational workflow, this view shows the inputs, outputs and integration characteristics already defined.

CapabilityIntegration summaryInputOutputBatchRealtimeMore
Loading integrations...

Developer reference

Technical information for callable capabilities

Endpoints, authentication, inputs, outputs, request formats and documentation for capabilities that have a developer-accessible implementation.

CapabilityEndpointMethodAuthVersionDocsMore
Loading developer docs...

Built around requirements

If the capability you need is not listed, that does not mean Oscar cannot build it

The catalogue grows as new requirements are implemented. The starting point is the information or outcome a client needs, not whether an identical capability already exists in the table.

A useful capability brief can be very simple

Tell us what information you want Oscar to retrieve or work with, what you need done to it, and what a useful result should look like. We can then determine the sources, logic, AI layer and access route.

1. What do you need to know or retrieve?Define the information need, source type, entity, market or question.
2. What should Oscar do with it?Retrieve, extract, match, validate, classify, compare, summarise, score, monitor or transform.
3. What should the result contain?Define the fields, answer, structure, confidence, summary or other useful output.
4. How should it be accessed?API, MCP-enabled route, integration, feed, PSN, application or bespoke delivery.

Have a public sector data or AI capability in mind?

Tell us the requirement. It may map to an existing capability or become the next one added to Oscar's catalogue.