System: Detection Engine — Build Pre-launch
Narrative Defense Platform · Crypto & Web3

When the talk is engineered, you find out last.

HorizonKey detects coordinated narrative attacks against crypto and Web3 projects, alerts your team in real time, and arms you with the response toolkit to fight back. Built for builders, not analysts.

01 · The Gap

Your monitoring stack tells you people are talking. It can't tell you when that talk is manufactured.

Coordinated Campaigns
20%

Average token price decline during a coordinated FUD campaign within 48 hours. Most projects identify the attack only after the bleed.

Capital Allocated, Wrong Door
$58M

Funding into enterprise narrative intelligence platforms targeting Fortune 500 and national security clients. None of it serves builders.

Defense Budget
$0

What most Web3 projects spend annually on narrative defense. Not because the threat isn't real — because the tooling doesn't exist for them.

02 · Detection Engine

Four phases. From noise to response, in minutes.

PHASE 01

Baseline.

We profile the normal conversation around your token. Volume distribution, sentiment spread, source typology, posting cadence. The baseline is your project's narrative fingerprint when nothing is wrong.

  • platformsx · reddit · news
  • windowrolling 30d
  • refreshhourly
PHASE 02

Detect.

Coordination signals: account age clustering, message similarity, posting time correlation, new-account ratio spikes. Behavioral pattern analysis runs through the LLM layer with structured threat scoring.

  • signals7 dimensions
  • target precision> 80%
  • analysis layerclaude api
PHASE 03

Alert.

Threat-scored, plain language, action-oriented. Slack, email, webhook. We tell you what's happening, who's behind it, and what to do — not a dashboard you have to interpret yourself.

  • channelsslack · email · webhook
  • latency< 5min
  • formatbrief + evidence pack
PHASE 04

Respond.

Detection alone is half the platform. The response toolkit ships in-product: platform-specific report templates, timestamped and hashed evidence packs for legal escalation, counter-narrative playbooks, and escalation workflow rules. Built in, not bolted on.

  • templatesx · reddit · youtube
  • evidencetimestamped + hashed
  • playbookscounter-narrative
SAMPLE · Conversation Volume — Coordinated Spike Detected BASELINE → ANOMALY → ALERT
T+0 · COORDINATION DETECTED PEAK · ALERT DISPATCHED
03 · Pricing

Self-serve. Fixed pricing. No demo call required.

Monitor is passive observation. Defense is active response. Two personas, two tiers — pick the one that matches the question you're trying to answer.

Monitor
$149/mo
"I want to see what's happening."
  • 1 brand monitored
  • X + Reddit coverage
  • Daily scan
  • Weekly digest
  • Email alerts
Defense Team
$499/mo
Team-scale defense for studios, DAOs, multi-token operators.
  • Unlimited brands
  • API access
  • Competitive monitoring
  • Multi-seat
  • Custom reporting
04 · Narrative Attack Report

Dispatches from the field.

Weekly case studies of real coordinated campaigns we've identified — anonymized, decomposed, and turned into pattern libraries. First issue ships at engine v1.

DISPATCH 001
Issue #1
Anatomy of a 48-hour FUD campaign against a mid-cap L2.
In Draft
DISPATCH 002
Issue #2
When the bots speak in dialect: linguistic fingerprints of paid amplification.
In Draft
DISPATCH 003
Issue #3
Reddit coordination: how subreddit timing windows betray the campaign.
Planned
DISPATCH 004
Issue #4
Counter-narrative playbooks: what works, what backfires, what to never post.
Planned
05 · Founders Cohort

The first 50 projects get founder access.

We're not launched. We're in detection-engine build. Join the waitlist and you'll be onboarded with founder pricing, direct access during build, and right of refusal on case studies featuring your data.

Request Access
Request Logged · We'll be in touch.

We don't share your data. We don't sell it. We don't subscribe you to anything you didn't sign up for.