AI Agents
Discovery Express

Knowledge

Research Analyst Agent

Answers the question your team would spend two days on, with every claim traced to its source.

In productionReviewed September 3, 2026
Slack / Teams · Research Analyst
Demo data

  1. Analyst

    How exposed is our top account to the new packaging rule in its market?

  2. Agent

    Two of their four product lines are in scope. That comes from their own filing and the rule text, both linked. What I could not find is their compliance timeline, so I have not guessed one.

  3. Brief written to SharePoint · 14 claims, 14 sources · 2 gaps listed · permissions of the requester applied

  4. Handoff to a person

    The analyst fills the two gaps with a client call. A claim without a retrievable source is never written.

Slack / TeamsSharePointConfluenceSnowflake
KPI it moves
Two days of desk research to two hoursMeasured
Channels
Slack / TeamsInternal appsEmail
Integrates
  • SharePoint
  • Confluence
  • Snowflake
  • Your CRM
  • Content sources
A research analyst reading printed reports with highlighted passages
Knowledge

What it does

Every answer sourced. Every claim traced back to where it came from.

Where it lives

Your cloud, your repo, your IP.

1. Receives 2. Looks up 3. Decides 4. Acts 5. Escalates 6. Learns

Channels

  • Slack / Teams
  • Internal apps
  • Email

Research Analyst Agent

03Decides

Systems it integrates

  • SharePoint
  • Confluence
  • Snowflake
  • Your CRM
  • Content sources

A person on your team

01

Receives

Takes the question in Slack or Teams and restates the scope before spending an hour on it.

02

Looks up

Reads your internal repositories, your data warehouse and the external sources you approved.

03

Decides

Separates what the sources support from what they do not, and says which is which.

04

Acts

Writes the brief with the answer first, every claim linked to its source and the gaps listed.

05

Escalates

Says out loud when the evidence is thin instead of filling the gap with a confident sentence.

06

Learns

Keeps the sources your analysts kept and drops the ones they always discard.

KPI it moves

Measured, with the source.

Every figure has a row in the Proof Ledger: source, date, owner and client permission.

Proof Ledger
Measured

Measured

2 days to 2 hours

Time to a sourced brief

Range observed across deployments of this agent, Sep 2026

Before

2 days

After

2 hours

Measured

100%

Claims with a traceable source

Enforced by the platform: a claim without a source is not written. Internal measurement, Sep 2026

Measured

75% to 90%

Briefs accepted without rework

Range observed across deployments of this agent, Sep 2026

Reviewed September 3, 2026

How it is governed

Every action has an owner, an eval and a limit.

Set in the Blueprint, enforced by the Govern layer of the platform.

Govern · Approval queue
Demo data

Waiting for a person

Brief for a client leaves the company · the author decides

14 claims, 14 sources

Weekly eval

Passing

Citation accuracy

99.3%

Last 8 runs
96.2% → 99.3%

It declines when the evidence is insufficient, as expected

Limit, enforced

2 unsourced claims removed from the draft

Human approvals

  • Any brief that leaves the company, by its author
  • The list of approved external sources

Typical evals

  • Citation accuracy: every claim checked against its source
  • Answer quality scored by the analysts each week

Limits

  • It never writes a claim without a source
  • Document permissions are respected per person

Time to production

8 to 12 weeks

of pilot to production under a KPI Contract, after a Blueprint of 1 to 2 weeks

The agent already runs in production elsewhere, so the 2 to 4 weeks of tuning inside the pilot go to your policies, your data and your systems, not to the model.

DaCodes delivery standard, Sep 2026. Integration is the long pole, not the model.

How we deploy
The four phases
Commitment gate

  1. 01Discovery Express

    2 sessions, no cost

  2. 02Blueprint

    1 to 2 weeks

  3. 03Pilot

    8 to 12 weeks

  4. 04Scale & Managed

    Continuous

  • Pilot, first 8 weeks
  • Up to week 12 of the pilot
  • This agent: 2 to 4 weeks of tuning, inside the pilot
  • Commitment gate: the KPI decides Scale

Proof and demo

What backs the label.

Proof Ledger note

In production

Measured in a real deployment and logged in the Proof Ledger with source, date and owner.

Reviewed September 3, 2026

Where this agent sits on the maturity policy

  1. Reference architecture
  2. Pilot
  3. In productionToday

The Discovery Express shows this agent on your data, live. Two sessions, no cost.

See this agent with your data

Related agents

Same function or industry.

All agents

FAQ

Direct answers.

A claim without a retrievable source is not written: the check runs in the Govern layer, and every citation is verified against the document it points to before the brief is delivered.

No. It inherits the permissions of the person who asked, so two people asking the same question get answers built from what each of them can already open.

Only the ones you approve, and it never sends confidential content out to reach them. The source list is versioned in your repo.

Which process do we automate first?

Two discovery sessions, no cost, and you leave with two or three quick wins whether or not you build with us.

We reply in under one business day.

Loading the form