Skip to main content
skill CX Codex

market-pattern — Claude Code skill: candlestick pattern detection

Scans ingested OHLCV data for candlestick patterns, classifies each by type and reliability, and stores the result in memory.

Install into Claude Code · Skills CLI (Vercel) All methods and platforms ↓
npx skills add ruvnet/ruflo --skill market-pattern -a claude-code
Open on GitHub

What market-pattern does

The skill walks candle sequences looking for familiar shapes: doji, hammer, inverted hammer, bullish and bearish engulfing, morning and evening star, three white soldiers, three black crows, plus head & shoulders, double top and bottom, and cup & handle. Every hit gets a name, a type (reversal or continuation), a direction, and a reliability score from 0.0 to 1.0. Results come back sorted by that score.

It does not collect data. Candles have to be ingested first through market-ingest, otherwise there is nothing to scan. Reads go through mcp__plugin_ruflo-core_ruflo__memory_search or memory_list on the market-data namespace: the memory_* family routes by namespace, while agentdb_hierarchical-* routes by tier, so namespaced reads need memory_*.

Storage has two paths. The typed one uses mcp__plugin_ruflo-core_ruflo__agentdb_pattern-store with type: 'market-pattern' and no namespace argument, since ReasoningBank handles routing; if the bridge is unavailable the fallback writes to the reserved pattern namespace with controller: 'memory-store-fallback'. The plain path is memory_store --namespace market-patterns, which does respect the namespace. The call takes a symbol and an optional period: <symbol> [--period 1D].

What it helps with

  • spot doji and hammer shapes in ingested candles
  • tell a reversal pattern from a continuation one
  • rank detected patterns by reliability score
  • store a pattern annotation under a key
  • review pattern history for one symbol and period

Example prompts for the agent

  • Find candlestick patterns for AAPL on the 1D period
  • Run market-pattern on BTC-USD and sort by reliability
  • Store the detected pattern in the market-patterns namespace

How to install market-pattern

1Skills CLI (Vercel)

npx skills add ruvnet/ruflo --skill market-pattern -a claude-code

2Manually: put the files in place

~/.claude/skills/market-pattern/SKILL.md

Copy the whole skill folder; for one project use .claude/skills/market-pattern/

Step by step, with commands for each platform: How to install skills in Claude Code: five ways

Questions about market-pattern

How do I install market-pattern in Claude Code?

Run npx skills add ruvnet/ruflo --skill market-pattern -a claude-code. The skill file lands at ~/.claude/skills/market-pattern/SKILL.md.

What does market-pattern do?

It scans ingested OHLCV data for candlestick patterns, assigns each one a type, direction and reliability score, sorts by reliability, and stores the result in memory.

Why does the skill find nothing?

Data must be ingested first via market-ingest. Without that, the market-data namespace holds no candles for the symbol and period.

Source text

SKILL.md · MIT
---
name: market-pattern
description: Detect and classify candlestick patterns from ingested OHLCV data
argument-hint: "<symbol> [--period 1D]"
allowed-tools: mcp__plugin_ruflo-core_ruflo__memory_search mcp__plugin_ruflo-core_ruflo__memory_list mcp__plugin_ruflo-core_ruflo__memory_store mcp__plugin_ruflo-core_ruflo__agentdb_pattern-store mcp__plugin_ruflo-core_ruflo__agentdb_pattern-search mcp__plugin_ruflo-core_ruflo__ruvllm_hnsw_route Bash
---

# Market Pattern

Scan ingested OHLCV data for known candlestick patterns, classify them by type and reliability, and store for future reference.

## When to use

When you need to identify candlestick patterns (doji, hammer, engulfing, head-shoulders, etc.) in market data. Requires data to be ingested first via `market-ingest`.

## Steps

1. **Load candles** -- call `mcp__plugin_ruflo-core_ruflo__memory_search` (or `memory_list`) on the `market-data` namespace to retrieve normalized OHLCV data for the symbol and period. The `memory_*` tool family routes by namespace; the `agentdb_hierarchical-*` family does NOT (it routes by tier), so use `memory_*` for namespaced reads.
2. **Scan for patterns** -- iterate through candle sequences looking for:
   - Single-candle: doji (open ~= close), hammer (long lower wick), inverted hammer
   - Two-candle: bullish/bearish engulfing
   - Three-candle: morning star, evening star, three white soldiers, three black crows
   - Multi-candle: head & shoulders, double top/bottom, cup & handle
3. **Classify** -- for each detection, assign: pattern name, type (reversal/continuation), direction (bullish/bearish), reliability score (0.0-1.0)
4. **Rank** -- sort by reliability score descending
5. **Store** -- two paths (per ruflo-cost-tracker ADR-0001 dual-path pattern):
   - **Pattern-store (typed, recommended)**: `mcp__plugin_ruflo-core_ruflo__agentdb_pattern-store` with `type: 'market-pattern'`. Don't pass a `namespace` arg — ReasoningBank routes it; on bridge unavailability the fallback writes to the reserved `pattern` namespace with `controller: 'memory-store-fallback'` (see ruflo-agentdb ADR-0001).
   - **Plain store (namespace-routable)**: `mcp__plugin_ruflo-core_ruflo__memory_store --namespace market-patterns` — this DOES respect the `market-patterns` namespace because `memory_*` is namespace-routed.
6. **Report** -- display: pattern name, date range, direction, reliability, suggested action

## CLI alternative

```bash
npx @claude-flow/cli@latest memory search --query "bullish reversal patterns" --namespace market-patterns
npx @claude-flow/cli@latest memory store --key "pattern-AAPL-2026-05-04-doji" --value '{...}' --namespace market-patterns
```

© ruvnet · MIT · ruvnet/ruflo

Similar skills

Everything for AI and ML →

From the same set146

workflow-create Skill to author workflows: MCP templates with pause-and-resume gates, or native JS scripts with subagent fan-out. For long pipelines and complex audits. CX Codex +1 738installs ddd-context Scaffolds a standard DDD bounded context directory structure with domain, application, and infrastructure layers. Useful when splitting monoliths. CX Codex +1 738installs adr-index Builds an index of architecture decisions plus the graph of links between them, all in one run of scripts/import.mjs. Useful when the ADR files on disk drifted from what the agent remembers. CX Codex +1 737installs trader-train Trains neural network models on market data using the neural-trader CLI. Prepares market forecasts with confidence intervals for specific tickers. CX Codex +1 736installs iot-register Registers a Cognitum Seed device by endpoint and sets up an agent bridge. Useful when wiring the device into Claude. CX Codex +1 736installs ddd-validate Checks bounded context integrity in TypeScript projects and flags domain boundary leaks. Useful for DDD architecture audits and pre-merge CI checks. CX Codex +1 736installs sparc-implement A skill that runs SPARC design phases: writes algorithm pseudocode, defines module boundaries and API contracts, then generates working code. Use it if you're developing with SPARC methodology and want to automate the architecture stage. CX Codex +1 734installs observe-metrics Aggregates agent swarm counters, gauges, and histograms over a time window to detect metric anomalies against baselines. CX Codex +1 733installs

Project sponsors

Companies that keep this analytics open
Gold sponsor

Impressions, clicks and code copies — open statistics. We keep the numbers public for the same reason NeBlask itself is public. Become a sponsor