# Qlib Factor Backtest

> Ingest market data for a supplied universe and backtest a supplied alpha-factor definition with Qlib. Use when a researcher wants a direct, evidence-linked backtest of one or more already-defined factors; this workflow does not discover new factors or place trades.

- Skill: `knuckles-team/qlib-factor-backtest` (Agent Skill, multi-file: 2 files)
- Install (CLI): `npx skillmds@latest add knuckles-team/qlib-factor-backtest`
- Raw SKILL.md: https://api.skillmd.com/api/skills/knuckles-team/qlib-factor-backtest/raw
- Safety review: pending
- Works with: Claude Code, Claude.ai, OpenAI Codex
- Category: Productivity
- License: MIT
- Author: Knuckles-Team (https://skillmd.com/u/knuckles-team)
- Updated: 2026-09-17
- Page: https://skillmd.com/skills/knuckles-team/qlib-factor-backtest

---


# Qlib Factor Backtest Workflow

Compose the named atomic skills without adding factor-discovery logic here.

## Inputs

Provide the factor definition(s), asset universe, and date range.

## Steps

### Step 0: quant-data-ingest [skill: quant-data-ingest]

Invoke `$quant-data-ingest` with the workflow inputs to ingest market
data for the universe and date range.

Expected: `normalized_market_dataset`

### Step 1: qlib-backtester [skill: qlib-backtester] [depends_on: Step 0]

Invoke `$qlib-backtester` with `normalized_market_dataset` and the
supplied factor definition(s).

Expected: `backtest_report`

## Output

Return `backtest_report`. Does not discover new factors or place trades.

## Execution

- **Run first:** Step 0 — `$quant-data-ingest`.
- **After level 0:** Step 1 — `$qlib-backtester`.

**Execution:** If graph-os is reachable, offload the whole DAG via `graph_orchestrate action=execute_workflow` (or the `kg-delegate` skill) for true parallel/swarm execution. Otherwise execute the steps natively in dependency order: run steps with no unmet `depends_on` in parallel, then their dependents.