| name | routine_cookbook |
| description | The single reference for writing Condor routines — fetching Hummingbot data, parallel calls, reports/charts, continuous loops, and candlestick charts. Routes to a companion file per topic. |
| when_to_use | Before implementing or debugging ANY routine. Read this first, then pull the specific companion file(s) for what your routine actually does (data, async, reports, continuous, charts). |
| source | agent:routine_builder |
Routine Cookbook
The patterns for building routines, split into companion files so you load
only what your task needs. Read this overview, then fetch the relevant file(s):
manage_skill(action="read_file", name="routine_cookbook", file="hummingbot_client.md")
Which companion file to read
| Your routine needs to… | Read |
|---|
| Fetch market data, candles, prices, order book, portfolio, executors | hummingbot_client.md |
| Make 4+ parallel API calls / bulk fetch many pairs / rate-limit | async_patterns.md |
| Produce a report — KPIs, tables, Plotly charts, rich inline output | report_builder.md |
| Run a continuous loop (monitor, tracker, alerts) until stopped | continuous.md |
| Render a candlestick chart, indicator overlay, or volume footprint | candles_chart.md |
Most routines need report_builder.md plus one or two others. A continuous
price monitor with a live dashboard, for example, reads hummingbot_client.md
Non-negotiables (apply to every routine)
- Every routine MUST generate a ReportBuilder report — see
report_builder.md.
- All client calls are async — always
await; never time.sleep, only asyncio.sleep.
- Parse defensively: handle
None/missing keys, return error strings, never raise to the caller.
- Test after writing (
manage_routines(action="run", ...)) and fix until the output is clean.
These are starting patterns, not a bypass — running a routine still goes through
the normal execution/confirmation controls.
