Plugins API

A plugin is a manifest (plugin.json) plus a Python entry point. It runs in a sandbox inside the engine: it may import a vetted set of modules, reach the network through urllib, and write only inside the data cache. A broken plugin never crashes the engine.

Manifest schema

textCopy{
  "id": "vix_data",
  "name": "VIX Data",
  "version": "1.0.0",
  "author": "You",
  "description": "...",
  "type": "data_source | indicator | strategy_template | exporter | ui_widget",
  "language": "python",
  "entryPoint": "main.py",
  "config": { "cache_days": { "type": "number", "default": 30, "label": "Cache TTL (days)" } },
  "dependencies": [],
  "tags": ["data", "vix"]
}

Lifecycle: register(api)

The entry point defines register(api). The api registers what the plugin provides:

• api.register_data_source(id, name, download_fn, config_schema)
• api.register_indicator(id, name, factory_fn, param_specs)
• api.register_strategy(id, name, strategy_class, param_specs)
• api.register_exporter(id, name, export_fn): export_fn(result), or export_fn(result, options) to receive the options the export was asked with. It runs when an export is asked for, with every file it writes kept inside a new folder for that one export. It returns the path of the file it wrote, or a text.
• api.register_data_fetch(fetch_fn) runs before each backtest (see External Data)
• api.get_config(key, default) reads a config value: the one saved in the plugin's settings in the Plugins view, else the manifest default
• api.log(message) appends to the plugin log shown in the Plugins view

VIX plugin walkthrough

The shipped VIX plugin is a data plugin built on the data_fetch hook.

• Config series_id (default "VIXCLS") selects the FRED series, so the same plugin can fetch any FRED series by changing one value.
• Config cache_days (default 30) is the cache TTL in days; the plugin refetches only when the cache is older than this.
• data_fetch(context) downloads the FRED CSV, parses it into a date to value map, and writes vix.json into context.data_dir.
• A strategy reads it with ctx.data("vix", date).

pythonCopydef on_market(self, ctx):
    bars = ctx.bars(1)
    if not bars:
        return
    vix = ctx.data("vix", bars[-1].ts)
    if vix and vix > 30:
        ctx.signal("EXIT")