FoundronicsGet your free score
Menu

AGENT · FOUND · BE THE ANSWER WHEN THEY LOOK

FOUND-U: the Underdog Moves agent

Answers the question: Have we made anything distinctive enough to be cited? Measured by: Original research, benchmarks, open tools, public data and their citation rate.

Score your Underdog Moves for freeRead the Underdog Moves letter
Agent spec · read from the FOUND-U specification

FOUND-UUnderdog Moves

In

  • Client profile, ICP, category terms and approved claims (from the str…
  • Previous week's FOUND-O handoff packet.

Does

  • Load inputs and verify evidence from the prior cycle is intact.
  • Measure the letter's metrics (Original research, benchmarks, open too…

Out

  • Metric snapshot for this letter with method, provider/model, date, pr…
  • Draft recommendations and drafted assets (fixes, outreach drafts, bri…

You approveStrategist approves decision-grade findings and anything published or sent externally.

Simulation · Example Cyber Co does not exist · 90 days

2 drafts from 3 findings

  • Published after approval2
  • Waiting for a person0
  • Stopped by the checks0

The spec card is read from the agent's specification. The run card is from a simulation on made-up data; no real company is shown.

  • Specification published, and this page is built from it.
  • Runs in the simulation on a fictional company Demonstration
  • Connects to your systems Not built yet

How it works

What it does, step by step

The agent prepares. The Strategist agent checks. You approve anything that leaves the company.

Where it stops and waits for you

  • Strategist approves decision-grade findings and anything published or sent externally.
  • Review requests, analyst briefings and outreach are sent only by a human or after explicit approval.
  • Agents never create or buy reviews, testimonials, analyst opinions, awards or other third-party validation.
  1. Load inputs and verify evidence from the prior cycle is intact.
  2. Measure the letter's metrics (Original research, benchmarks, open tools, public data and their citation rate) using deterministic calculation; LLMs may classify and draft rationales only.
  3. Compare to last cycle and flag deltas and anomalies.
  4. Draft the fixes or assets for the gaps found.
  5. Package findings with citations and confidence.
  6. Submit to the strategist for approval, then hand off.

Hands on: To: FOUND-N Nods. Packet: metric snapshot, gap list, approved assets, open questions, evidence links.

What goes in, and what comes out

Inputs

  • Client profile, ICP, category terms and approved claims (from the strategist's scoping record).
  • Previous week's FOUND-O handoff packet.
  • Prior-cycle metrics for this letter (never overwritten, append only).
  • Sanctioned sources only (official model APIs, public pages, listings); no scraping of consumer AI interfaces.

Outputs

  • Metric snapshot for this letter with method, provider/model, date, prompt version and sources recorded.
  • Draft recommendations and drafted assets (fixes, outreach drafts, briefs) marked `draft`.
  • Evidence log entries; handoff packet for the next stage.

Simulation · fictional data

What it prepared for Example Cyber Co

Example Cyber Co does not exist. These are drafts from a 90-day simulation, not work for any real company. See the whole simulation.

  • Day 1 · Published after approval

    Structured data (JSON-LD) for the answer pages

  • Day 1 · Published after approval

    Brief: an original survey of security practitioners

What Underdog Moves means, how it is scored, the harness the agent runs in and what to avoid are on the letter page, so they are kept in one place.

Read the Underdog Moves letterSee its harnessLearn FOUND-U free in the AcademyNext agent: FOUND-N NodsAll twelve agents

See how Underdog Moves looks for your company today.

Free, open and no login. Every number shows who saw the evidence.

Get your free FOUND ScoreGet your improvement plan
Get your free FOUND Score