# Academic Alpha Scanner

> Discover academic market-factor research, audit its evidence, extract comparable hypotheses, obtain the required market data, and backtest the candidates by composing package-owned and universal atomic skills. Use when a quantitative researcher wants an evidence-linked screening report for candidate alpha factors, not live trading or investment advice.

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

---


# Academic Alpha Scanner Workflow

Compose the package-owned ScholarX skill and named universal atomic skills. Keep
research findings and backtest results clearly separate from investment advice or
live-order execution.

## Inputs

Provide the research question, markets, asset universe, date range, source filters,
candidate-selection criteria, data constraints, and backtest assumptions.

## Steps

### Step 0: scholarx-operations [skill: scholarx-operations]

Invoke `$scholarx-operations` with the workflow inputs.

Expected: `paper_source_packet`

### Step 1: citation-auditor [skill: citation-auditor] [depends_on: Step 0]

Invoke `$citation-auditor` with `paper_source_packet`.

Expected: `evidence_audit`

### Step 2: factor-hypothesis-extractor [skill: factor-hypothesis-extractor] [depends_on: Step 1]

Invoke `$factor-hypothesis-extractor` with `paper_source_packet`, `evidence_audit`,
and the workflow inputs.

Expected: `candidate_factor_hypotheses`

### Step 3: quant-data-ingest [skill: quant-data-ingest] [depends_on: Step 2]

Invoke `$quant-data-ingest` with `candidate_factor_hypotheses` and the workflow
inputs.

Expected: `normalized_market_dataset`

### Step 4: qlib-backtester [skill: qlib-backtester] [depends_on: Step 3]

Invoke `$qlib-backtester` with `candidate_factor_hypotheses`,
`normalized_market_dataset`, and the workflow inputs.

Expected: `backtest_report`

## Output

Return `paper_source_packet`, `evidence_audit`, `candidate_factor_hypotheses`, and
`backtest_report`, including the assumptions and limitations emitted by each skill.
Do not place or recommend live trades.

## Execution

- **Run first:** Step 0 — `$scholarx-operations`.
- **After Step 0:** Step 1 — `$citation-auditor`.
- **After Step 1:** Step 2 — `$factor-hypothesis-extractor`.
- **After Step 2:** Step 3 — `$quant-data-ingest`.
- **After Step 3:** Step 4 — `$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.