deepKPI Financial Analysis and Research Tools

Query structured KPIs extracted from SEC filings for US public companies. Powered by Revelata.

This skill suite covers different aspects of financial analysis workflows. This includes data access via MCP for Claude or REST API for OpenClaw, data retrieval, metric derivation, seasonality analysis, and Excel export. Sub-skills are organized so that each layer can be loaded independently — only read what the task requires.

Data access — Claude vs. OpenClaw

The deepKPI operations (query_company_id, list_kpis, search_kpis, company_summary_search, get_company_summary, get_company_segments, list_sec_filing_markdowns, get_sec_filing_markdown) are available two ways depending on the runtime:

retrieve-kpi-data/retrieve-kpi-data.md is the KPI metrics workflow; retrieve-sec-filing/retrieve-sec-filing.md is the SEC filing markdown workflow; company-summary-segments/company-summary-segments.md is what companies do (summary), segment / geo structure, and thematic discovery (get_company_summary, get_company_segments, company_summary_search). The DEEPKPI_API_KEY env var and deepkpi-api reference doc are only needed in OpenClaw (or as an env-var fallback when MCP is unavailable).

Hard stop: deepKPI connection failures

If deepKPI access fails (MCP connector not working, REST calls failing, auth errors, network issues, or you cannot retrieve data), you MUST STOP and ask the user:

“I can’t access deepKPI right now. Do you want to proceed without deepKPI?”

• If the user says no, stop.
• If the user says yes, you may proceed using other sources (e.g. web), BUT you MUST NOT use deepKPI skill branding, templates, formatting conventions, or “Powered by Revelata deepKPI” framing for non-deepKPI data. Clearly label alternate sources.

This rule applies to ALL sub-skills in this bundle.

Sub-skill routing

Read the relevant reference doc(s) before doing any work. Multiple docs may apply to a single request — load all that are relevant.

Default entry point: For metrics and modeling feeds, start with retrieve-kpi-data/retrieve-kpi-data.md (it references derive-implied-metric, analyze-seasonality, and format-deepkpi-for-excel as needed). For business description, segment structure, or thematic company lists, start with company-summary-segments/company-summary-segments.md. For filing markdown (full or excerpts) and verbatim quotes, start with retrieve-sec-filing/retrieve-sec-filing.md. If a request mixes these, read every doc that applies.

Sub-skill summary

deepkpi-api — Raw REST access to deepKPI endpoints: query_company_id, list_kpis, search_kpis, company_summary_search, get_company_summary, get_company_segments, list_sec_filing_markdowns, get_sec_filing_markdown. OpenClaw / env-var fallback only — in Claude, use the native MCP tools instead.

retrieve-kpi-data — Primary workflow for structured KPIs (company ID, list_kpis / search_kpis, gap handling including Q4 derivation), provenance rules, in-chat tables, and the mandatory post-pull Excel offer.

company-summary-segments — get_company_summary (what they do) and get_company_segments (how they break out segments / geos); company_summary_search for thematic “who does X?” discovery. Not for KPI time series.

retrieve-sec-filing — SEC filing markdown in chat: full filing or excerpts, verbatim quotes, MD&A / risk language. Tools: list_sec_filing_markdowns → get_sec_filing_markdown, source hierarchy (tool-first, not web), blockquote discipline for quote-style asks. Distinct purpose from KPI time series.

derive-implied-metric — Compute metrics that deepKPI doesn't report directly from data that IS reported: Q4 = FY − (Q1+Q2+Q3), missing segment = total − known segments, per-unit economics (revenue per store, ARPU, ASP), take rates, penetration rates, and geographic mix percentages.

analyze-seasonality — Compute each quarter's typical share of the full fiscal year from 2–3 years of actuals, then apply those ratios to split an annual projection into quarterly estimates.

format-deepkpi-for-excel — Canonical layout and styling spec for any .xlsx export built from deepKPI data: wide layout with periods as columns (annual block → blank column → quarterly block), Calibri font, green input cells, clickable hyperlinks in cells, numeric date headers, freeze panes, column grouping, and no redundant Source rows.

analyst-report-pressure-test — End-to-end workflow for uploaded analyst PDFs: extract 4–6 claims, pull granular deepKPI metrics per claim, build paired "supports" / "complicates" evidence with mandatory provenance hyperlinks, and emit a Revelata-branded interactive HTML report (references/html-template.md, references/chart-patterns.md). After the draft HTML exists, run the mandatory internal draft HTML QA rubric in analyst-report-pressure-test/pressure-test-double-check.md, then revise and re-check as needed before presenting the file. Always read retrieve-kpi-data for KPI mechanics; use retrieve-sec-filing when the test needs verbatim filing passages; use company-summary-segments for business description / segment structure / thematic lists when useful; use derive-implied-metric when filling Q4 or segment gaps.

peer-benchmark — Curated operational benchmark sets (not trading multiples): company_summary_search for candidates, KPI fingerprint + alignment, diff-driven segment sub-benchmarks, chat 1-pager, optional HTML (references/html-template.md), Excel via format-deepkpi-for-excel. Always read retrieve-kpi-data first.

idea-generation-survey — Onramp when the user has no ticker and no thesis: rapid interview (multiple-choice by default), company_summary_search for a 10–15 name screen with thesis-matched KPIs, then deep pulls on 1–2 chosen names per retrieve-kpi-data. references/sector-kpi-map.md for diagnostic KPI concepts. No default Excel offer. Not for users who already named a company or a specific metric — use retrieve-kpi-data instead.