Non-reliance 8-Ks, auditor changes, impairments, and XBRL fundamentals — screened in plain English and cited to the filing, for audit, advisory, and forensic work.
One protocol — bring it to the assistant you already use
The same jobs, without the exports, the tabs, or the unverifiable answer.
Before
Restatement risk surfaced when the news wrote about it.
With MCP
Screen for Item 4.02 non-reliance and Item 4.01 auditor changes the day they are filed.
Before
Comparables meant pulling XBRL by hand or paying for another terminal seat.
With MCP
Quarterly fundamentals for ~4,400 filers back to ~2009 — growth, margins, capex, leases — in one question.
Before
Finding the language meant reading the whole filing.
With MCP
Pull the exact sentences behind a signal — impairment, going concern, control weakness — each cited.
Run these today in Claude, ChatGPT, or any MCP client — each returns a cited answer.
Which companies filed non-reliance 8-Ks (Item 4.02) last quarter?
Show auditor changes in the software sector this year.
Pull NVDA's last 8 quarters of revenue, margins, and capex intensity.
Find going-concern language in recent 10-Ks for this peer set.
Connect any subset — each is a focused, cited surface for its own domain.
The data is derived from public SEC filings and every figure cites its source, so anything you take forward is independently verifiable at the filing.
No. Nothing you hold is uploaded or ingested, and your queries are not used to train models.
Filings flow through the pipeline as the SEC publishes them; news refreshes daily.
Values come from company-filed XBRL, quarterly back to ~2009, with sector-specific fields surfaced only when the company actually reported them — nothing imputed.
Public-filing data only — no client data ingested, no training on queries, read-only tools. How we handle data
Request a demo and we'll walk through the workflow above with your names, your dates, and your questions.
Request a Demo