eva_timeseries
Financial time series
Trace one filing-backed metric across periods, including growth rates, CAGR, and a chart.
Try: “Chart Microsoft revenue for the last five fiscal years and calculate CAGR.”
EVA Connector
Connect Claude, ChatGPT, Microsoft Copilot, or another MCP host to EVA’s read-only financial research tools.
The EVA connector is a standard Model Context Protocol (MCP) connection between your AI assistant and Ridhics EVA. It gives the host a focused catalog of read-only tools for SEC filings, XBRL financials, earnings-call research, comparison, provenance, and engine-gated valuation. You can ask in ordinary language; the host selects the appropriate EVA tool and returns the structured result in the conversation.
EVA is built to keep the research trail visible. Supported answers carry receipts that identify source facts, filings, periods, accessions, or calculation inputs, with deep links back to SEC material where available. That lets you move from a summary to the underlying evidence instead of treating an uncited model answer as the final record.
Fourteen public research tools, presented consistently across supported connector hosts.
eva_timeseries
Trace one filing-backed metric across periods, including growth rates, CAGR, and a chart.
Try: “Chart Microsoft revenue for the last five fiscal years and calculate CAGR.”
eva_statements
Open a complete income statement, balance sheet, or cash-flow statement for one period.
Try: “Show Apple’s latest annual income statement with source links.”
eva_metrics
Review a compact dashboard of scale, margin, return, leverage, cash-flow, and growth metrics.
Try: “Summarize NVIDIA’s latest margins, returns, leverage, and cash-flow metrics.”
eva_segments
Break out filing-backed results by product, geography, or reportable segment.
Try: “Break down Amazon’s latest annual revenue by reported segment.”
eva_multiples
Compare peer valuation multiples with coverage and availability diagnostics.
Try: “Compare Microsoft with its peers on P/E, EV/EBITDA, EV/Sales, and free-cash-flow yield.”
eva_compare
Compare the same filing-backed financial metric across multiple public companies.
Try: “Compare Apple, Microsoft, and Alphabet on their latest annual operating margins.”
eva_ask_eva
Ask narrative research questions and receive grounded answers with filing or transcript citations.
Try: “What did NVIDIA management say about data-center demand on its latest earnings call?”
eva_filing_search
Find exact, verbatim passages in SEC filings with accession-scoped source metadata.
Try: “Find Apple’s exact latest 10-K language about supply-chain risk.”
xbrl_facts
Retrieve one or a few exact as-filed XBRL values for a company, metric, and period.
Try: “What was Apple’s as-filed FY2025 revenue?”
eva_dcf
Run an engine-gated intrinsic DCF, valuation cases, or valuation range with explicit assumptions.
Try: “Run a base-case DCF for Microsoft and show the assumptions and calculation receipt.”
eva_reverse_dcf
Solve for the operating assumptions implied by the current market price.
Try: “What revenue growth and margin assumptions are priced into NVIDIA?”
eva_audit_trail
Inspect a figure’s source fact, SEC accession, restatement chain, and calculation lineage.
Try: “Audit Apple’s FY2025 operating cash flow back to the source filing.”
eva_filing_metadata
List filing forms, dates, accession numbers, fiscal periods, and source links.
Try: “List Microsoft’s latest 10-K and 10-Q filings with SEC links.”
eva_catalog
Explore supported metrics, coverage, valuation methods, examples, and provenance capabilities.
Try: “What valuation methods and provenance checks does EVA support?”
Use the same EVA endpoint in each host: https://mcp.ridhics.com/mcp
EVA uses standard MCP. Other clients that support authenticated remote MCP servers can connect to the same endpoint and complete the same OAuth sign-in flow.
EVA grounds structured results in SEC filings and XBRL facts and grounds earnings-call research in transcript evidence. Receipts and deep links expose the filing, accession, fiscal period, source fact, transcript passage, or calculation inputs used. Where a workflow distinguishes as-filed facts from as-reported presentation, EVA labels that choice. When required evidence is missing, stale, contradictory, or outside a supported coverage gate, EVA is designed to abstain honestly and name the failed check rather than fill the gap with a guess.
Use the connector as a research aid, with its coverage and decision boundaries visible.
Structured financial tools focus on U.S. SEC filers. Valuation availability is engine-gated by company and method, so an unsupported request returns a named limitation rather than an invented value.
Connector tools retrieve research data and calculations. They do not place trades, move money, modify filings, or write back into your source systems.
EVA does not provide personalized financial advice, buy or sell recommendations, guaranteed returns, or individualized portfolio instructions. Usage allowances and response rates depend on the applicable EVA plan and host.
Authentication and research access are designed around least privilege and traceability.
Users sign in through EVA; hosts do not receive or store the user’s EVA password.
Organization-scoped access is reinforced with database row-level security boundaries.
Security-relevant connector activity is recorded for operational review and accountability.
See security, privacy, data-handling, and procurement details.
Visit the Trust CenterFor setup or account help, email support@ridhics.com.