mirror of
https://github.com/OpenBB-finance/OpenBB.git
synced 2026-10-27 12:40:35 +00:00
d3d33a219b
* refactor openbb-technical for v5 * forward refs * linting * no python 3.15 in test matrix * reame * ty * move all params to body --------- Co-authored-by: deeleeramone <>
2.6 KiB
Vendored
2.6 KiB
Vendored
openbb-technical Documentation
The technical extension exposes ~60 endpoints organised into four families:
| Family | What it does | URL prefix |
|---|---|---|
indicators/* |
Single-symbol overlays, oscillators, trend, volume, volatility, structure, statistics | / (root) |
signals/* |
Event detectors: crossovers, divergences, breakouts, candlestick patterns, regime | /signals/ |
multi/* |
Multi-symbol correlation, screening, relative-rotation, multi-indicator composition, catalog | / (root, except correlation_matrix and screen which expose multi.*) |
relative_rotation |
JdK RS-Ratio + RS-Momentum for a benchmark-anchored basket | / (root) |
Every endpoint is reachable through three transports:
- Python:
from openbb import obb; obb.technical.<endpoint>(...) - HTTP:
POST http://localhost:8000/api/v1/technical/<endpoint>(runopenbb-apito start the server) - CLI:
openbb -ithen/technical/<endpoint>
Where to start
- First time? getting-started.md shows installation, the three transports, and a 5-line example for each.
- Building a strategy? user-guides/01-indicators.md covers single-symbol indicator workflows.
- Want trade signals? user-guides/02-signals.md covers the signals subrouter.
- Working with a basket? user-guides/03-multi-symbol.md covers correlation, screening, and Relative Rotation Graphs.
- Composing pipelines? user-guides/04-composition-and-discovery.md covers the
multiandindicators(catalog) endpoints. - Looking up a specific endpoint? api-reference.md lists every endpoint with its parameters, output columns, and a one-line description.
Conventions
- Input: every data-consuming endpoint accepts the same shapes —
list[Data],list[dict],pandas.DataFrame,pandas.Series, ornumpy.ndarray. The library converts between them automatically. - Output: every endpoint returns an
OBBject. The numeric payload is on.resultsas a list of typed Pydantic models. - Validation: every endpoint validates its inputs through a
XxxQueryParamsPydantic class and emits its rows through aXxxData(orXxxEvent/XxxSignal) class. The class docstrings carry full NumPy-formatParametersandReturnssections. - Discovery: the
indicatorsendpoint introspects the registered router and returns a structured catalog — useful for building UI pickers, OpenAPI clients, or LLM tool-calls without hard-coding endpoint names.