Examples API reference What only Genka has
Examples
Composed flows organised by what you're trying to do. Pick the workflow that matches your job, not the data source. Each page walks one real research session end to end — tools, curl, expected output.
export GENKA_API_KEY=imk_live_... H="X-API-Key: $GENKA_API_KEY"
Workflows by job-to-be-done
Equity research →
One stock, four lenses. Concall narrative
(concall_search) → valuation trajectory
(co_ratios) → peer cohort
(sector_get) → institutional positioning
(/companies/{X}/holders). HINDALCO worked
through end to end.
Portfolio review →
What you already own, examined for duplication, concentration
risk, valuation drift, stale data. Holdings overlap before
consolidating two funds. Reverse-index for total exposure to a
single stock across the book. Stale-data guards on the disclosure
as_of.
Macro watching →
Tape vs tone. FII / DII flow across cash + F&O — the
fii_opt_30d row at -₹83K cr is the actionable detail.
Cross-reference against concall search filtered to macro themes:
US rate cuts, China demand, premiumisation. When tape and tone
agree, conviction is high.
Discovery →
From a vague theme to a defendable shortlist. Federated
search for entities tied to a keyword, narrow with
mf_screener or co_screener, confirm the
thesis is alive in management voice. Negative results are
information — if no fund matches your theme, the vehicle
doesn't exist yet.
By data source (deep references)
The pages below are organised by corpus rather than workflow — useful when you already know which dataset you want and need the full surface.
Concalls →
35+ ranked queries against the 17,979-doc corpus. Capex, margins, AI capex, ethanol, NIM, channel destocking.
Mutual funds →
Holdings overlap, subscription gates, peer-by-holdings, survivorship-clean backtest, XIRR.
FII / DII flow →
Daily cash + F&O participant breakdown, 5d/30d rolling, 30-day cumulative time-series with Nifty context.