Introducing
VIGIL™
Proactive CAT Briefing & Action.
Watching. Before the storm. Before the claim. Before the loss.
VIGIL
/ ˈvɪdʒɪl /
Latin — vigilia · "watchfulness"
A period of watchful attention maintained before an expected event. The act of staying alert and ready to act — especially through a period of danger or uncertainty. Used historically by sentinels who watched through the night so others could act at dawn.
Where VIGIL sits in the CAT Sentinel ecosystem
Detects
CAT Sentinel
Real-time NWS alerts · 14 perils · 12 data layers · underwriting signal engine
→
Records
CAT Event Book™
Living library · verified event intelligence · mitigation ROI · peril trend atlas
→
NEW
Acts
VIGIL™
Proactive briefings · automated action plans · role-specific workflows · before the event hits
The inspiration: what OpenClaw does — and how VIGIL goes further
OpenClaw approach
AI-generated briefings from live data feeds
✓ Monitors data sources continuously
✓ Generates plain-language summaries
✓ Surfaces key findings for executives
— General-purpose · not domain-specific
— No action routing to specific roles
— No insurance / CAT domain intelligence
VIGIL™ — CAT-native extension
Proactive briefings + routed action plans for insurance
✓ Monitors CAT Sentinel + Event Book continuously
✓ Generates role-specific briefings (UW, Claims, CRO)
✓ Routes specific action items to specific teams
✓ Grounded in 14-peril CAT domain intelligence
✓ Tracks action completion and re-escalates
✓ Pre-event · not post-loss
Before 24h
Average briefing lead time before event peak
3 roles
Underwriting · Claims · Executive — each briefed differently
100% routed
Every action item assigned to a specific owner and deadline
Always on
Continuous monitoring · no manual trigger required
What is VIGIL™
The proactive intelligence layer that turns CAT signals into coordinated action — before the loss occurs.

VIGIL™ monitors the CAT Sentinel signal feed and the Event Book™ continuously. When conditions meet a configurable threshold, it automatically generates a structured briefing and routes specific action items to the right role — underwriter, claims director, or CRO — with a deadline and an owner.

01
Continuous Signal Monitoring
VIGIL™ watches CAT Sentinel's 14-peril alert feed, the Event Book™ frequency trend data, and carrier portfolio accumulation in real-time. No manual checking required.
02
Threshold Detection & Trigger
When a configurable threshold is crossed — e.g. 3+ simultaneous HIGH alerts in a state, or a peril trend gap exceeding 10pp — VIGIL™ triggers a briefing cycle automatically.
03
Role-Aware Briefing Generation
One event — three briefings. VIGIL™ generates a CRO executive summary, an underwriting action brief, and a claims pre-positioning brief simultaneously. Each calibrated to its audience.
04
Action Routing & Assignment
Every briefing ends with a set of specific, time-bound action items. VIGIL™ routes each to the right owner — re-rate to underwriting, adjuster staging to claims, treaty review to risk management.
05
Completion Tracking & Re-escalation
VIGIL™ tracks whether actions are completed before the event. Overdue items re-escalate automatically. After the event, it generates a post-event comparison: what was briefed vs. what happened.
Signal Monitoring Dashboard
Texas SCS Alert Feed
168 HIGH · Active
CA WLF Frequency Trend
+12pp gap detected
LA HUR+FLD Accumulation
$24B TIV · Watch
FL Portfolio Accumulation
Within threshold
VIGIL™ monitoring
Live
Threshold Configuration
SCS Trigger Rule
IF alerts ≥ 5 states simultaneously
AND any state CAT index ≥ 80
→ CRITICAL briefing
Pricing Gap Trigger
IF observed trend gap > 10pp
AND renewal cycle < 90 days
→ HIGH briefing
Accumulation Trigger
IF 3 events within 28 days
AND same 3-state corridor
→ TREATY briefing
Three simultaneous briefings — one event
📊
CRO Executive Brief
Portfolio exposure · treaty adequacy · board implications
📋
Underwriting Action Brief
Re-rate triggers · deductible changes · new business holds
🚁
Claims Pre-Position Brief
Adjuster staging · vendor activation · timing curve
🔄
Reinsurance Treaty Alert
Per-occurrence attachment test · aggregate accumulation
Action routing example — TX SCS critical event
→ Underwriting Team · Due 48h
Apply SCS frequency surcharge +18–22% on TX renewals due < 90 days
→ Reinsurance · Due 72h
Run 1-in-10yr SCS scenario against per-occurrence attachment
→ Claims Director · Due 24h
Stage 40 adjusters in Harris + Dallas counties ahead of forecast peak
→ CRO · For information
Portfolio SOS: $68B TIV in active HIGH-alert zone. 3 actions in progress.
Action tracker — post-briefing
SCS surcharge applied
✓ Done · 31h
40 adjusters pre-staged
✓ Done · 22h
Treaty attachment test
⚠ Overdue · 4h
Re-escalated to CRO
Auto-escalated
Post-event brief scheduled: compare briefed scenario vs actual claims in 14 days
Briefing Engine
A sample VIGIL™ briefing — generated automatically

This is what VIGIL™ generates when Texas crosses a critical SCS accumulation threshold. Three versions produced simultaneously — CRO, Underwriting, Claims. Below is the underwriting brief.

VIGIL™ Underwriting Brief
Critical
Auto-generated · 06:14 CST
📍 Texas — Statewide
⚡ SCS + FLD
Event active
👤 Underwriting team
Situation Summary
CAT Sentinel has recorded 168 HIGH-severity SCS alerts in Texas over the past 9 months — one event every 1.6 days. Three simultaneous HIGH-severity events are active across Harris, Dallas, and Travis counties as of this briefing. Combined TIV exposure in active alert zones: $68 billion.
The CAT Event Book™ Peril Trend Atlas shows an observed SCS frequency trend of 18%/yr annualised — versus the 5% standard model assumption embedded in current pricing. This gap has been compounding for 3+ renewal cycles. The current portfolio is structurally underpriced for this peril in this state.
Required Actions — Priority Order
🔴
Immediate: SCS frequency surcharge — TX renewals <90 days
Apply +18–22pp surcharge to all Texas SCS-exposed renewals due within 90 days. This corrects the 13pp observed-vs-model gap. Do not wait for next scheduled pricing review.
Owner: TX Underwriting Lead · Deadline: 48 hours
🔴
Wind/hail deductible: separate from AOP immediately
All Texas homeowner policies must carry a separate wind/hail deductible of 1–2% of dwelling value. AOP-only deductibles leave the book exposed to the frequency pattern now documented in the Event Book.
Owner: Product Management · Deadline: Next renewal cycle
🟡
New business hold: Harris County flood-zone properties
54% protection gap in Texas means the insured pool is adversely selected. Pause new flood-zone business in Harris County pending updated adverse selection load calculation.
Owner: Underwriting Director · Deadline: 72 hours
🔵
Treaty attachment: run 1-in-10yr SCS scenario
March 2025 SCS outbreak tested 6 states simultaneously at $9.5B. Run your per-occurrence attachment against this scenario before the next treaty renewal.
Owner: Reinsurance Manager · Deadline: 5 business days
Intelligence Sources
CAT Sentinel · TX signal register Event Book™ · Peril Trend Atlas Event Book™ · County Risk Index Protection Gap Map · TX 54%
Three briefing formats — one trigger
CRO Executive Brief · 1 page
Portfolio exposure headline + board-ready narrative
Total TIV at risk · treaty adequacy status · 3 actions in progress · estimated net loss scenario. No technical detail — decision-ready language only.
Underwriting Action Brief · shown above
Pricing gap + risk selection + treaty actions
Specific surcharge percentages · deductible changes · new business holds · treaty test scenarios. All grounded in Event Book™ data with source citations.
Claims Pre-Position Brief · ops-ready
Adjuster staging + vendor activation + timing curve
How many adjusters · where to stage · which vendors to activate · when pipe-freeze claims will peak. Derived from Event Book™ peril timing data.
Post-Event Debrief · auto-generated at T+14
What VIGIL™ briefed vs what actually happened
Side-by-side: briefed scenario vs actual claim count, actual loss estimate vs prediction. Feeds back into Event Book™ to improve future briefings.
Delivery channels
📧 Email 💬 Slack / Teams 📱 Push notification 🖥 In-platform dashboard 🔌 Webhook / API 📄 PDF export
Technical Architecture
How VIGIL™ is built on top of CAT Sentinel

VIGIL™ is a composable layer that sits above the existing CAT Sentinel signal infrastructure. It adds a monitoring engine, a briefing generation layer (powered by the Anthropic API), and an action routing system — without replacing anything already in place.

Five-stage processing pipeline
📡
Signal Ingest
CAT Sentinel alert feed · Event Book™ data stream · portfolio accumulation data
⚙️
Rule Engine
Configurable thresholds · peril-specific triggers · portfolio-specific rules
✦
Briefing Gen
Claude API · role-aware prompts · Event Book™ context · 3 simultaneous briefs
🎯
Action Router
Role mapping · deadline assignment · owner notification · escalation rules
📊
Tracking Loop
Completion status · re-escalation · post-event debrief · Event Book™ feedback
Data sources VIGIL™ draws from
CAT Sentinel
Real-time Signal Feed
14-peril HIGH-alert stream · 12 data layers · state risk indices · portfolio accumulation · underwriting signal register. The primary trigger source for VIGIL™ briefing cycles.
CAT Event Book™
Intelligence Context
Peril Trend Atlas for observed vs. model gaps · County Risk Index for TIV context · Siege Register for accumulation history · Mitigation ROI for action recommendations.
Carrier Systems
Portfolio Context
Policy-level TIV by county · renewal calendar · claims handler rosters · vendor contracts. Connected via secure API. VIGIL™ never stores underlying policy data.
✦
Powered by Claude (Anthropic API)
The same AI that powers CAT Sentinel's insights engine — now generating proactive briefings
VIGIL™ uses the Anthropic Claude API with a CAT-domain system prompt grounded in the Event Book™ intelligence base. Every briefing is generated from verified CAT data — not from general internet knowledge. The system prompt evolves as the Event Book™ grows, meaning VIGIL™ briefings improve continuously alongside the data.
Use Cases
Six scenarios where VIGIL™ acts before the loss

Each use case shows the trigger condition, the briefing generated, and the action taken — all before a claim is filed.

SCS / Severe Convective Storm
Texas accumulation — 5-state outbreak
TRIGGER: 6 states simultaneously active · CAT index TX 92 · $68B TIV alert zone
UW: SCS surcharge +18–22% on TX renewals <90 days
Claims: Stage 40+ adjusters in Harris + Dallas pre-event
Re: Run 1-in-10yr per-occurrence attachment test
CRO: Portfolio SOS — $68B at risk, 3 actions in progress
WLF / Wildfire
California January wildfire — off-season trigger
TRIGGER: WLF HIGH alert in January · 5 months outside model window · $89B TIV CA
UW: Convert WLF pricing from 6-month to 12-month immediately
UW: No new policies within 500m active WUI perimeter
UW: WUI defensible space audit at all CA renewals
Re: CA WLF 1-in-10yr scenario = $12–20B · test attachment
FLD / Flood
Mississippi 72% protection gap — adverse selection signal
TRIGGER: MS protection gap 72% · FLD HIGH alert active · poverty 19.8%
UW: Adverse selection load +18–22% on all MS FLD-zone policies
UW: Mandatory FLD endorsement or declination — 100yr zone
Claims: Reserve uplift — 18–24mo MS settlement window assumption
Reg: Monitor MS mandatory flood bill — active in legislature
WIN / Winter Storm
Pennsylvania 3-storm accumulation window
TRIGGER: 3 HIGH WIN events in 28 days · PA pipe-freeze timing curve active
Claims: Pipe-freeze claims peak in 72h post-thaw — adjusters ready
Re: 3-event 28-day window — aggregate vs per-occurrence definition check
UW: PA accumulation window limit review for next renewal cycle
CRO: Two separate claims waves expected — PA siege register entry
HUR / Hurricane
Florida Gulf Coast — pre-season treaty test
TRIGGER: NOAA above-normal season forecast · FL CAT index 91 · $68B TIV
Re: 1-in-10yr FL HUR scenario $18–25B — test per-occ before June
UW: Named-storm deductible 2–5% DV — enforce before season opens
UW: AOB clause restriction — all coastal FL renewals this cycle
Claims: Demand surge +30–35% load · vendor contracts pre-locked
Portfolio-wide
Annual pricing cycle — model assumption review
TRIGGER: Annual renewal cycle opens · observed trend gap >10pp in 4+ perils
Actuarial: Re-weight trend load to observed rates for SCS, WLF, HUR, FLD
UW: All books priced on pre-2020 data — flag for mandatory review
Product: Credibility-weight to most recent 5-yr window — not 30yr
CRO: 3–4× model underpricing compounding — board briefing required
The Investment Thesis
The insurance industry is reactive. VIGIL™ makes it proactive.

Every major CAT loss in 2025 — $40B LA wildfires, TX Hill Country floods, PA winter accumulation — had a detectable signal in the data 12 to 72 hours before peak loss. The industry had the data. It didn't have VIGIL™.

→
CAT Sentinel detects the signal
Real-time 14-peril NWS feed · 12 data layers · underwriting signal register
→
Event Book™ provides the context
What this peril did in the past · what it means for pricing · what mitigation works
→
VIGIL™ generates the action
Role-specific briefing · routed action items · deadline · owner · completion tracking
VIGIL™ vs. existing approaches
Capability VIGIL™ Manual process Generic AI tool
CAT-domain intelligence✓ Native✗~ General only
Proactive — no manual trigger✓ Always on✗ Human dependent✗
Role-specific briefings✓ CRO / UW / Claims✗✗
Action routing with deadlines✓ Auto-assigned~ Manual✗
Event Book™ grounding✓ Every brief✗✗
Post-event debrief loop✓ Auto at T+14~ Ad-hoc✗
Development Roadmap
1
Phase 1 · Q3 2025
VIGIL™ Core — Briefing Engine
Rule-based trigger engine, Claude API briefing generation, 3 role templates (CRO, UW, Claims), email + Slack delivery. Integrated with CAT Sentinel signal feed.
2
Phase 2 · Q1 2026
Action Router + Tracking
Action item routing to named owners, deadline management, overdue re-escalation, completion dashboard. Integration with carrier workflow systems via webhook.
3
Phase 3 · Q3 2026
Post-Event Debrief Loop
Auto-generated T+14 debrief comparing briefed scenario vs actual loss data. Carrier-shared loss signals feed back into Event Book™ to improve future briefings. Continuous learning loop.
4
Phase 4 · 2027+
VIGIL™ Standalone Product
Available independently of CAT Sentinel for carriers using their own underwriting platform. Connects to carrier data systems and the Event Book™ Data API.
Why VIGIL™ is a defensible moat
Domain intelligence is the barrier
Any tool can generate a briefing. Only VIGIL™ generates a briefing grounded in the CAT Sentinel signal register and the Event Book™ — giving it a specificity no general AI tool can replicate.
The feedback loop compounds the advantage
Every post-event debrief makes the Event Book™ richer. A richer Event Book™ makes VIGIL™ briefings more accurate. The moat widens with every event the system processes.
Action routing creates operational stickiness
Once underwriting, claims, and risk management teams have their workflows routed through VIGIL™, the product becomes embedded in how the organisation operates — not a dashboard anyone can switch off.
Standalone addressable market is 10× larger
Phase 4 makes VIGIL™ available to carriers not on CAT Sentinel. The proactive briefing market is the entire insurance industry — not just CAT Sentinel subscribers.
Start the conversation
The industry had the data.
Now it has VIGIL™.

Proactive CAT briefings. Role-specific action plans. Continuous monitoring. Grounded in the CAT Sentinel signal register and the CAT Event Book™ — the only AI briefing engine built exclusively for catastrophe risk.