Status: VALIDATION IN PROGRESS. The system architecture, canon, portable scan skill, scanner/plugin surface, and failure taxonomy are built. Full clean end-to-end acceptance remains a deployment gate before this should be presented as a finished production system. The scan results shown in the cover image are illustrative, not live trade recommendations.
The problem
Most options scanners optimize for finding something. That creates exactly the wrong incentive: stale quotes, weak liquidity, missing catalysts, hand-waved probability, or a provider failure can still end in a confident-looking trade idea.
I wanted the opposite behavior. If the evidence is bad, the scanner should fail loudly. If the market scan is incomplete, it should say incomplete. If nothing survives the rules, it should return no match.
What I built
VERA Options Scanner / Options Income System is a separate opportunity-research and contract-selection framework for options structures with expirations of 30 DTE or less:
Canon-first execution so current strategy rules override stale assumptions from older sessions
A portable options-income-opportunity-scan skill that standardizes request parsing, market-data retrieval, probability filtering, structure construction, risk filtering, liquidity review, technical analysis, catalyst analysis, viability gating, ranking, and result formatting
Broad discovery / pagination logic that applies cheap filters before expensive options-chain calls
A 70%+ probability-of-expiring-OTM target where that metric is meaningful, using direct provider probability when available or a clearly labeled delta-based proxy
Required liquidity checks including bid, ask, midpoint, spread width, volume, open interest, and quote / Greeks freshness
Required technical checks on the underlying, and required catalyst / earnings checks through expiration
Prompt-specific monetary limits instead of silently inventing a permanent account budget or risk percentage
Explicit separation from the main Macro Trading portfolio and its state
The reliability rule I care about most
The scanner uses different states for different outcomes:
QUALIFIED_CANDIDATE — the scan completed and something actually passed
NO_MATCH — the scan completed and nothing passed
SCAN_INCOMPLETE — valid work happened, but the requested universe/process was not exhausted
DATA_UNAVAILABLE — required market or research evidence could not be obtained
SCAN_FAILED — infrastructure, provider, or tool failure prevented the requested scan
A provider quota error is not “NO TRADE.” A connector answering is not success if the requested scan did not complete.
What I can do for a client
Design an AI-assisted market or opportunity scanner with explicit acceptance/rejection gates
Build research workflows that combine structured data, technical checks, catalysts, liquidity, and risk constraints
Add failure-state semantics so outages, stale data, partial scans, and genuine “nothing qualified” results cannot be confused
Design provider-budget-aware pagination and staged filtering for expensive APIs
Separate research authority from execution authority so an AI scanner cannot quietly become a trading bot
Attribution: Phillip Wells owns problem framing, system/workflow design, constraints, evaluation standards, failure diagnosis, revision decisions, and documentation. AI platforms provided implementation and analytical assistance under that direction.
I built an options-research system that searches for defined opportunities under explicit rules — and is required to admit when data or infrastructure failed instead of dressing a broken scan up as “no trade.”