Research Agent
Evidence and citation specialist
Collects evidence from approved sources, compares findings side by side and produces citation-backed research you can audit.
A sample-data demonstration workflow is available. Live execution needs more configuration.
What this agent is for
Replace unsourced summaries with a traceable evidence table where every claim carries the source it came from.
Capabilities
- Source collection from approved repositories
- Claim-by-claim evidence comparison
- Citation-backed synthesis
- Contradiction and gap detection
- Executive research brief
Inputs it needs from you
- Research question and scope
- Allowed source list
- Reference documents
Outputs it produces
- Evidence table with citations
- Comparison matrix
- Research brief
- Open questions list
Industries
- Government
- Education
- Healthcare
- Energy
- Professional services
How This AI Agent Works
The published process for this agent: what triggers it, what it validates, which approved tools it calls, where a human approves and what it delivers.
Evidence-based research workflow
From a research question to a cited, human-reviewed report.
Saved with the run so the recorded execution trail shows what was asked. Demonstration runs use sample data. They do not touch your systems.
Public visitors can preview the structure and inspect every node. Labelled demonstrations and live runs require a signed-in account with configured providers.
Workflow summary as a text list
- 1. Research question: Trigger. A user submits the question, scope and required depth.
- 2. Clarify scope: AI agent. Restate the question, list assumptions and set inclusion criteria.
- 3. Search approved sources: Approved tool. Query only the source list the workspace permits.
- 4. Retrieve evidence: Data source. Collect passages, publication dates and source identifiers.
- 5. Compare sources: AI agent. Group agreements, contradictions and gaps.
- 6. Validate citations: Validation. Verify every claim maps to a retrievable citation.
- 7. Human review: Human approval. A reviewer confirms the evidence before release. Human approval required.
- 8. Cited research report: Output. Findings, comparison table and full citation list.
Approval gate: Human review. Deliverable: Report.
Workflow audit for this page
| Template ID | Nodes | Approval node | Output type | Preview | Demo | Live | Mobile test |
|---|---|---|---|---|---|---|---|
| research-agent | 8 | Yes. Policy: always | Report | Available | Available (signed in) | Available via Studio · requires ai_provider | Verified at 390px |
Demonstration runs are stored and replayed from recorded events. Live runs execute server-side in the Studio and only start when the listed integrations report healthy.
Ask this agent team a question
Ask the live agents
Describe the problem in your own words. We classify the industry, assign specialist agents, and you watch the workflow run and produce the visual deliverables — maps, dashboards, charts, tables and reports — instead of one block of text.
Demo mode uses clearly labelled sample data — never presented as verified.
Assigned specialists
Industry detected from your wording
- Requirement validation agent
- Analysis agent
- Quality and citation reviewer
Deliverables planned
Workflow the run follows
- 1. User request
- 2. Requirement validation
- 3. Industry classification
- 4. Specialist agent selection
- 5. Data and file collection
- 6. Tool authorisation
- 7. Execution plan
- 8. Multi-agent collaboration
- 9. Analysis or generation
- 10. Quality validation
- 11. Human approval
- 12. Visual deliverables
- 13. Monitoring and audit
Ready to scope your AI initiative?
Share your objective and constraints. A specialist reviews every request before any recommendation is issued.