Gather

Analyst-Driven Research Corpus Assembly

Build the evidence base as the question evolves.

Gives analysts a connected workbench to pull from external sources — filings, papers, databases, web — and clip evidence directly against a live research question. The question can change as understanding deepens; the corpus grows with it, with full provenance on what was captured, from where, and why.

Shape

Operational dimensions

Human co-pilot

Person and system work side-by-side.

On demand

Fires when a user asks.

Medium data gravity

Holds working state that compounds over runs.

Read-only inbound

Consumes external data; does not write back.

Inputs

Research question or claim nodes defining the inquiry frame
Source connectors (academic databases, regulatory filings, web, vendor feeds)
Analyst capture actions (clip, tag, annotate, attach-to-claim)
Existing corpus items to avoid duplication

Outputs

Question-bound research corpus (collected evidence items)
Evidence-to-claim linkage graph with capture provenance
Capture trail (source, date, analyst, rationale per item)
Corpus coverage signal (gaps relative to current question frame)

Mechanism

Provides analysts an evidence-collection workbench against an evolving research question — query and pull from external sources, clip and capture passages, attach evidence to question/claim nodes, and grow a question-specific corpus over the life of the inquiry.

Why this is a primitive

Cannot be decomposed — the active-analyst-mediated pull-and-clip-against-an-evolving-question loop is one operation. Distinct from monitoring (autonomous, fixed query) because the question itself evolves with the analyst's understanding. Distinct from document-intake (push from external senders) because here the analyst pulls. Distinct from organise (the corpus is being grown, not structured post-hoc). The mechanism IS analyst-mediated active capture.

Where it shows up

M&A due diligence team — analysts build a deal-specific evidence corpus from filings, press, court records, and expert commentary against a structured diligence question set

Policy research unit — assembles evidence across academic papers, government reports, and stakeholder submissions for a specific policy question, maintaining claim-level evidence linkage

Competitive intelligence analyst — builds a corpus on a competitor's strategic moves (job postings, patents, pricing) against an evolving set of hypotheses

Litigation support team — collects and clips relevant precedent, regulatory guidance, and expert literature against specific claim elements in a case

Related primitives

Composes with

External Stream Monitoring →Live Conversation Capture →Inbound Document Intake →

Alternative to

External Stream Monitoring →