market-ingest

Ingest and normalize market data into OHLCV vectors with HNSW indexing.

70.1k|8.4k|Updated Jun 2, 2025
One-click install
npx skills add https://github.com/ruvnet/claude-flow --skill market-ingest
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: market-ingest
Source: https://github.com/ruvnet/claude-flow/tree/main/plugins/ruflo-market-data/skills/market-ingest
Command: npx skills add https://github.com/ruvnet/claude-flow --skill market-ingest

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Raw market data arrives in inconsistent formats and cannot be searched for similar patterns. This Skill fetches price and volume data for a symbol, normalizes it into comparable OHLCV vectors, and indexes them for fast nearest-neighbor pattern search.

Core Features & Use Cases

  • Data Normalization: Converts raw prices into relative OHLCV values and volume into Z-scores against rolling statistics, making candles comparable across time periods and symbols.
  • Vector Storage and Indexing: Persists normalized data in a namespaced memory store and adds 64-dimension vectors to an HNSW index for similarity search.
  • Use Case: Before running pattern detection on a stock symbol, ingest its historical candles so downstream analysis can query the HNSW index for candles with similar price and volume shapes.

Quick Start

Ingest market data for symbol AAPL and prepare it for pattern detection and similarity search.

Frequently Asked Questions about market-ingest

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I ingest market data for pattern detection?

Fetch OHLCV data for the symbol from a REST API, CSV file, or manual input, then normalize prices to relative values and volume to Z-scores. Store the result in the market-data namespace and add each candle vector to the HNSW index for nearest-neighbor search.

How to normalize OHLCV data for vector similarity search?

Convert open, high, low, and close to relative changes against the previous close or open, and compute volume as a Z-score against rolling mean and standard deviation. Each candle is then encoded as a 64-dimension padded vector.

What is HNSW indexing used for in market data analysis?

HNSW (Hierarchical Navigable Small World) indexing enables fast approximate nearest-neighbor search over candle vectors. It lets you find historical candles with similar price and volume patterns without scanning the entire dataset.

Can I ingest market data from a CSV file instead of an API?

Yes, the skill supports REST API, CSV file, or manual input as data sources. After loading, the same normalization, storage, and HNSW indexing steps apply regardless of the source.

Why use memory_store instead of agentdb tools for market data?

The memory_* tool family routes data by namespace, which fits the market-data namespace used for symbol and date keys. The agentdb_hierarchical-* family routes by tier and ignores namespace strings, so it is not suitable here.