Two Members Rated Critical
Of 83 members of Congress who received data-driven investigation dossiers from CapitolExposed's analysis engine, two triggered critical-severity ratings: Rep. Jefferson Shreve (R-IN) and Rep. Byron Donalds (R-FL). Critical severity is the highest classification in the system. It means multiple independent data sources converged on patterns that, taken together, represent the densest concentration of potential conflict indicators in the database.
The dossiers are not opinion documents. They contain no subjective assessments, no editorial commentary, and no allegations of wrongdoing. Each dossier is a structured output of cross-referencing a member's publicly disclosed financial transactions against seven independent databases. The engine checks for correlations that a human analyst would need weeks to identify manually. The system completes the analysis in minutes.
The critical rating for Shreve and Donalds does not mean they broke any law. It means the volume and pattern of cross-references connecting their financial activity to their legislative duties exceeded the thresholds that distinguish routine political activity from statistically unusual concentrations of potential conflict.
The Seven Analysis Components
CapitolExposed's investigation engine runs seven distinct analysis modules against each member's profile.
Committee overlap analysis compares every stock trade a member has made against the list of companies and sectors under the jurisdiction of their committee assignments. A member of the House Energy and Commerce Committee who trades pharmaceutical stocks triggers this module. The system pulls committee rosters from Congress.gov and maps traded companies to sectors using SEC Standard Industrial Classification codes.
Timing proximity analysis measures the distance between a member's trade dates and dates of committee votes, floor votes, and legislative markups involving the same sector. The system uses the congressional calendar and bill tracking data to identify legislative events, then calculates the gap in days between each trade and the nearest relevant event.
Lobbying connection analysis cross-references traded companies against the Lobbying Disclosure Act database. If a company the member traded filed LD-2 lobbying reports targeting the member's committee during the same congressional session, the system records the connection. The data comes from the Senate Office of Public Records.
Donation pattern analysis checks FEC records for campaign contributions from the traded company's employees, its political action committee, or its industry-aligned PACs to the trading member's campaign committee. The system records the total contribution amount, the timing of contributions relative to trades, and whether the contributions came from individual employees or from PAC accounts.
Trade size analysis compares each transaction against the member's own baseline. The system calculates a rolling 12-month average transaction size for each member and flags trades that exceed two standard deviations from that baseline.
Cross-reference density analysis measures how many connections a member has across four external databases: the International Consortium of Investigative Journalists offshore leaks database, the Foreign Agents Registration Act filings with the Department of Justice, the Financial Crimes Enforcement Network records, and the OpenSanctions consolidated watchlist. This module does not assess whether any individual connection is illicit. It counts the number of links and compares the density to the congressional baseline.
Sanctions exposure analysis checks whether the member, the member's donors, or the member's traded companies appear on international sanctions lists, counter-sanctions lists, or politically exposed persons databases maintained by OpenSanctions.
2,057 Cross-References
Across the 83 dossiers, the system identified 2,057 total cross-references spanning four independent databases. The ICIJ offshore leaks database, which contains records from the Panama Papers, Paradise Papers, Pandora Papers, and other major leaks, produced the largest number of hits. These hits do not mean the members themselves appear in offshore records. In most cases, the connection runs through companies traded by the member that also appear in the ICIJ database, or through donors to the member's campaign who are named in the leaks.
FARA filings produced the second-largest category of cross-references. The Foreign Agents Registration Act requires individuals and entities acting on behalf of foreign governments or foreign political parties to register with the Department of Justice. When a member of Congress receives campaign donations from individuals associated with a FARA-registered entity, or trades stock in a company whose lobbyists are FARA-registered, the system records the connection.
FinCEN records contributed additional cross-references, primarily through Suspicious Activity Reports and Currency Transaction Reports associated with entities connected to members' financial networks. OpenSanctions data provided the broadest coverage, drawing from over 100 official sanctions lists, regulatory actions, and law enforcement datasets worldwide.
The 2,057 cross-references are not evenly distributed. Some members have dozens of connections across multiple databases. Others have only one or two. The distribution follows a power law: a small number of members account for a disproportionate share of the total cross-references.
296 Members on Counter-Sanctions Lists
One of the more striking outputs of the investigation engine is that 296 members of Congress appear on foreign counter-sanctions lists compiled by countries including Russia, China, and Belarus. These counter-sanctions lists were created by foreign governments in response to U.S. sanctions and typically target American officials, legislators, and political figures as a retaliatory measure.
Appearing on a foreign counter-sanctions list is not an indicator of wrongdoing. It is a political act by a foreign government. Russia's counter-sanctions list, for example, includes most senior U.S. legislators who voted for Russia-related sanctions packages. China's list targets members who have been publicly critical of Chinese government policies.
The investigation engine records these appearances because they create potential complications for members who also have financial connections to entities operating in the sanctioning countries. A member who appears on Russia's counter-sanctions list and also trades stock in a company with significant Russian revenue exposure presents a different risk profile than a member with no such intersection.
Notable Dossiers
Three dossiers illustrate the range of findings the system produces.
Rep. Lisa McClain (R-MI) has 1,390 trades in the database, the highest volume among members with completed dossiers. Her dossier contains five distinct findings spanning trade_conflict, donation_pattern, and lobbying_link categories. The volume of her trading activity means the system has more data points to analyze, which increases the probability of identifying patterns. McClain's trades span multiple sectors, several of which fall under the jurisdiction of committees she has served on. Her donation records show contributions from industry PACs aligned with the sectors she trades most frequently.
Rep. Gilbert Cisneros (D-CA) has 910 trades and six findings, the highest count of any single dossier. Cisneros, a former lottery winner who entered politics with a personal fortune, trades at a volume and frequency that exceeds most members in either chamber. His six findings cover trade_conflict, donation_pattern, lobbying_link, offshore_connection, disclosure_gap, and contract_conflict categories. The breadth of categories means the cross-referencing engine found connections across nearly every database it checks.
Former Speaker Nancy Pelosi (D-CA) has 38 trades in the database and four findings. Pelosi's lower trade count reflects her practice of attributing trades to her husband, Paul Pelosi, whose transactions have drawn sustained public attention since 2020. Her dossier findings cover trade_conflict and donation_pattern categories. The Pelosi dossier is notable less for its severity score and more for the fact that the trades attributed to her household have been among the most publicly scrutinized in congressional history, and the system's automated analysis independently surfaced the same patterns that journalists identified through manual investigation.
Finding Categories Explained
The investigation engine classifies findings into six categories.
Trade_conflict is the most common finding. It fires when a member trades a stock with a direct connection to their committee jurisdiction, legislative activity, or regulatory authority. This is the core conflict-of-interest indicator that the STOCK Act was designed to address through disclosure.
Donation_pattern fires when the system identifies a correlation between campaign contributions and trading activity. The system looks for temporal patterns (donations and trades occurring in the same period), sectoral patterns (donations from the same industry as traded stocks), and directional patterns (donation increases preceding or following large trades in the donor's sector).
Lobbying_link fires when a traded company has active lobbying filings targeting the member's committee. The system does not assess whether the lobbying was successful or whether the member was personally involved in the lobbied legislation. It records the structural connection.
Offshore_connection fires when a cross-reference to the ICIJ offshore leaks database exists through a traded company, a donor, or a business associate. The connection can be indirect: a company the member traded may have a subsidiary that appears in the Panama Papers, for example.
Disclosure_gap fires when the member's trading disclosure timeline shows patterns suggesting late filing, retroactive disclosure, or gaps between transaction dates and filing dates that exceed the STOCK Act's requirements.
Contract_conflict fires when a member trades stock in a company that holds federal government contracts overseen by the member's committee. The system cross-references the Federal Procurement Data System with the member's committee assignments and trade records.
Methodology
The dossier system is deterministic. Given the same input data, it produces the same output every time. There is no AI-generated narrative, no language model interpretation, and no subjective weighting based on political factors. The system applies the same rules to every member regardless of party, chamber, seniority, or public profile.
The data sources are all public records. Trade data comes from Periodic Transaction Reports. Committee assignments come from Congress.gov. Lobbying data comes from LD-2 filings. Donation data comes from the FEC. Offshore data comes from the ICIJ Offshore Leaks Database. FARA data comes from the Department of Justice. Sanctions data comes from OpenSanctions. Government contract data comes from the Federal Procurement Data System.
The system does not have access to brokerage records, private communications, text messages, or any non-public information. It works exclusively with what the government and international transparency organizations have made available to the public. The dossiers represent what a thorough analyst could find by manually searching every available public database. The investigation engine does it systematically, at scale, and without the selection bias that inevitably affects human-directed investigations.
Eighty-three members now have completed dossiers. The engine is designed to update continuously as new disclosure filings, lobbying reports, and donation records become available. Each new data point triggers a re-evaluation of affected dossiers, and severity ratings adjust automatically based on the updated cross-reference density.
What the Dossiers Do Not Do
The dossiers do not allege criminal conduct. They do not recommend prosecution. They do not assess intent, motive, or subjective state of mind. They do not distinguish between a member who intentionally trades on inside information and a member whose legitimate portfolio management happens to correlate with their committee assignments.
The dossiers surface patterns. They quantify connections that exist in the public record but are difficult to see without automated cross-referencing. They provide a structured, standardized framework for evaluating the financial activity of elected officials against the backdrop of their official duties.
What happens with those patterns is a question for voters, journalists, ethics investigators, and law enforcement. CapitolExposed's role is to make the patterns visible. The investigation engine does not render verdicts. It opens files.