About
Built by people who have read a lot of dockets.
TrollerBk has compiled US business bankruptcy data since 2005. Investors, claims traders, law firms, auctioneers, and financial institutions have used that corpus for custom research for more than a decade.
What we do
Event feed, then the case file
We take the court event feed: sales, plans, disclosure statements, distributions, claim transfers. Each line is linked to a case, a docket, and a type. The product is a custom report around your mandate: we get you what you need, any way you want it. Do not rely on a prebuilt tool that does not fit. The portal is for sitting with a case once the list has named it.
We also build and run BankruptcyObserver.com, the sibling product for monitoring specific counterparties, and we collaborate with clients on additional datasets, including NursingHomeDatabase.com.
Contact
TexHost, LLC7113 San Pedro, #296
San Antonio, TX 78217
General and subscriptions: [email protected]
MCP and integration support: [email protected]
Enterprise inquiries: [email protected]
Start a conversationElsewhere
- MCP developer reference
- llms.txt for AI retrieval systems
- Terms of Service and Privacy
- BankruptcyObserver.com
How the corpus is built
Data practices
Sources. United States Bankruptcy Court records via PACER and the courts' public RSS feeds, and public claims registers from court-appointed claims agents. Every derived field links back to the original court document. See coverage methodology for outbound links to PACER and uscourts.gov.
Methodology. Business cases are non-individual debtors in the 94 US bankruptcy districts. You can open a docket in any of them. A few less-active courts do not publish an event feed we can harvest, so those filings do not appear on the national docket-event lists until someone opens the case. Docket events we do harvest are classified from the courts' own docket text. Coverage counts on this site are recomputed daily and rounded down.
Editorial. The public docket events sample table is generated from court docket text, not written or summarized by an AI, and published on a 30-day delay. Nothing on this website is produced by a model at page-load time. If you find an error in our data, tell us at [email protected] and we will correct the record.
Your information. We collect what we need to run your subscription and bill you, and we use it for nothing else. We do not sell subscriber data. Payment cards are handled by Stripe and never touch our servers. Details are in our Terms and Privacy.
Security. All traffic is encrypted in transit. MCP access uses OAuth 2.1 with PKCE and short-lived tokens, or a per-account API token you can recycle at any time from the portal.
Run the last 30 days against your mandate.
We will run the last 30 days of filings against your mandate before the first call.