# Alternative Data Pipeline

> Discover, fetch, and normalize non-traditional (web/news-sourced) market-data signals by composing the catalog's research and finance atomic skills. Use when a quant researcher needs alternative data ingested and quality-checked before using it in a factor or backtest pipeline; this workflow does not place trades or score signal predictiveness.

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

---


# Alternative Data Pipeline Workflow

Compose the named atomic skills without adding data-vendor logic here.

## Inputs

Provide the target instruments/universe, alt-data source categories (news, filings,
web-published datasets), lookback window, and any excluded sources.

## Steps

### Step 0: web-search [skill: web-search]

Invoke `$web-search` with the workflow inputs to locate candidate
alternative-data sources.

Expected: `alt_data_source_packet`

### Step 1: web-crawler [skill: web-crawler] [depends_on: Step 0]

Invoke `$web-crawler` with `alt_data_source_packet` to fetch the full
content of the discovered sources.

Expected: `raw_alt_data_documents`

### Step 2: quant-data-ingest [skill: quant-data-ingest] [depends_on: Step 1]

Invoke `$quant-data-ingest` with `raw_alt_data_documents` and the
workflow inputs to normalize records into the Timeseries Memory backend.

Expected: `normalized_alt_dataset`

### Step 3: data-quality-auditor [skill: data-quality-auditor] [depends_on: Step 2]

Invoke `$data-quality-auditor` with `normalized_alt_dataset` and the
declared rule set.

Expected: `quality_report`

## Output

Return `normalized_alt_dataset` and `quality_report`. Do not place or recommend
trades from this data alone.

## Execution

- **Run first:** Step 0 — `$web-search`.
- **After level 0:** Step 1 — `$web-crawler`.
- **After level 1:** Step 2 — `$quant-data-ingest`.
- **After level 2:** Step 3 — `$data-quality-auditor`.

**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.