AGENTICTRUTH
❯open veil-institutional-flow
← Type "back" or click [back-to-logbook] to return
evidence/veil-institutional-flow
EXHIBIT CACTIVE

Veil: Institutional Flow Intelligence

Detected 12 pre-earnings accumulations, 70% mobile engagement

Next.jsPythonPostgreSQLTailwindVercel
FINDINGProblem Statement
Retail investors lack access to institutional order flow data. Existing tools either too complex or actively misleading with predictions.
CONSTRAINTSBoundaries
▸Real-time data (15min delayed flow is useless)
▸No predictions — evidence only
▸Mobile-first UX (target audience trades on phone)
▸Must cross-reference 3+ data sources per signal
▸Zero false-positive insider alerts
APPROACHMethod
Built signal fusion platform: congressional trading disclosures, options flow, volume anomalies. Focused on "what happened" not "what will happen." No predictions, just evidence. Mobile-first design drove adoption.
RESULTSMeasured outcomes
12
Accumulation Patterns Detected
Pre-earnings, verified post-hoc
8%
Avg User-Reported Gain
On tracked positions
0
False-Positive Insider Alerts
Zero in production
70%
Mobile Engagement
Of total sessions
runbook-excerpt.log
❯ veil-ctl signals --last 24h
[09:15:33] INFO  Flow alert: NVDA unusual call volume (+340% vs 20d avg)
[09:15:34] INFO  Congressional disclosure: Sen. X purchased calls (filed 02-08)
[09:15:35] INFO  Signal confidence: HIGH (2/3 sources aligned)
[09:15:36] INFO  Alert delivered to 1,247 subscribers
[ OK ] 4 signals processed. 1 HIGH confidence.
LESSONSWhat we learned
01Show evidence, not predictions — users make decisions.
02Data latency matters: 15min delayed flow is useless.
03Mobile UX determines adoption for finance tools.
04Institutional data requires institutional-grade verification.
→ start conversation← back to console
~❯

Veil: Institutional Flow Intelligence

Status: ACTIVE

Outcome: Detected 12 pre-earnings accumulations, 70% mobile engagement

Problem

Retail investors lack access to institutional order flow data. Existing tools either too complex or actively misleading with predictions.

Constraints

  • Real-time data (15min delayed flow is useless)
  • No predictions — evidence only
  • Mobile-first UX (target audience trades on phone)
  • Must cross-reference 3+ data sources per signal
  • Zero false-positive insider alerts

Approach

Built signal fusion platform: congressional trading disclosures, options flow, volume anomalies. Focused on "what happened" not "what will happen." No predictions, just evidence. Mobile-first design drove adoption.

Results

  • Accumulation Patterns Detected: 12 — Pre-earnings, verified post-hoc
  • Avg User-Reported Gain: 8% — On tracked positions
  • False-Positive Insider Alerts: 0 — Zero in production
  • Mobile Engagement: 70% — Of total sessions

Lessons Learned

  • Show evidence, not predictions — users make decisions.
  • Data latency matters: 15min delayed flow is useless.
  • Mobile UX determines adoption for finance tools.
  • Institutional data requires institutional-grade verification.

Technology Stack

Next.js, Python, PostgreSQL, Tailwind, Vercel