chart-patterns
SkillCommerce & financeChart pattern recognition, price action patterns, candlestick signal bars, pullback bar counting H1/H2/H3/H4, trend channels, micro channels, trading ranges, breakouts, major trend reversals 5-step sequence, wedges, double tops and bottoms, triangles, head and shoulders, climaxes, measured moves, support and resistance
Available today. Use it from your connected AI after setup.
No other account needed.
Connect ahel once, and every AI you use reads what you have installed.
Then ask your AI: use the chart-patterns skill
What this skill tells your AI
The instructions your AI receives, as published by agentii-ai/agentii-investment-intelligence in plugins/vertical-plugins/technical-analysis/skills/agentii/chart-patterns/SKILL.md and read by ahel’s review.
Methodology inspired by publicly taught price action trading frameworks; all text is an original paraphrase.
Defaults
| Parameter | Default Value | Rationale |
|---|---|---|
| bar_count_threshold | 20 bars | Pullback > 20 bars = full Trading Range |
| reversal_steps | 5 | All trend reversals follow the same 5-step mechanical sequence |
| wedge_legs | 3 | Wedge requires three pushes at a trend extreme |
| climax_aftermath | Trading Range | After a climax, market enters TR before any reversal |
Preflight
Run canonical pre-flight per contracts/preflight.md. Propagate X-Agentii-Trace per contracts/x-agentii-trace-header.md.
Data Source Priority
- Chart patterns framework —
references/chart-patterns-reference.md(bundled full pattern methodology) - Upstream context — market structure classification from
price-actionskill output - Technical setups —
search_technical_setups(pattern_type=..., market_condition=..., timeframe=...)for matching againstgold.technical_setups
Methodology
Retrieval Scope
structured_only
Retrieval Strategy
This skill follows Branch (d) Simple Lookup from contracts/retrieval.md: the pattern recognition framework is bundled in references/chart-patterns-reference.md. Real-time price data via get_realtime_quote. Matched technical setups via search_technical_setups. No unstructured document retrieval.
Temporal Scope
See frontmatter temporal_scope block.
Tool Allowlist
See frontmatter allowed_tools.
Protocol
This skill implements the price action trading pattern recognition framework: 17 distinct chart pattern types, each with specific identification criteria, decision rules, and entry/exit parameters. The core architecture: Setup = Context + Signal Bar — context (from price-action skill) determines which patterns are valid. A perfect signal bar in the wrong context is a losing trade. Detailed methodology and pattern-specific rules are in references/chart-patterns-reference.md.
Step 1 — Macro Context (Mode C — degrade-gracefully)
- Assess VIX via
get_realtime_quotefor broad market proxy. - VIX elevated (> 25) → tighten position sizing, favor defined-risk setups. Low VIX (< 15) → standard sizing.
- Degrade gracefully if unavailable: annotate
coverage_gapand proceed.
Step 2 — Context from price-action Skill
Receive or derive market structure classification:
market_condition: trending_up / trending_down / rangingtrend_strength: strong / weak(channel) / nonealways_in_direction: long / short / neutralcycle_phase: breakout / channel / trading_range / reversal
Critical rule: Context determines which patterns are valid:
- Strong bull trend + always_in_long → only long patterns (H1, H2, bull breakouts). All short patterns invalid.
- Strong bear trend + always_in_short → only short patterns (L1, L2, bear breakouts).
- Ranging → Buy Low/Sell High only near boundaries. Middle of range = no trade.
Step 3 — Pattern Identification
Apply the full 17-pattern recognition framework based on current context:
If in a Trend:
- Pullback depth → count legs: 1-leg = H1/L1 (strong trend), 2-leg = H2/L2 (channel), 3-leg = H3/L3 (broader), 4-leg = H4/L4 (near TR).
- Channel boundaries → identify trend line and channel line. Assess tight vs broad.
- Breakout signals → true vs false breakout assessment using bar characteristics.
- Reversal watch → check if 5-step reversal sequence is progressing (Steps 1-5).
If in a Trading Range:
- Range boundaries → identify support and resistance levels from prior swing points.
- Breakout preparation → monitor for breakout direction.
- TR size assessment → small TR (< 20 bars) = pullback, direction biased. Large TR (≥ 20 bars) = direction neutral.
If near a Trend Extreme:
- Climax check → consecutive strong bars → sudden large counter-bar → climax detected. After climax: expect TR, not immediate reversal.
- Wedge check → three pushes to same area, each push weaker → exhaustion signal.
- 5-step reversal → if Steps 1-4 are complete, watch for Step 5 confirmation.
- Double top/bottom → two tests of same level; neckline break for confirmation.
Step 4 — Setup Matching via MCP
Match identified patterns against the gold.technical_setups database:
search_technical_setups(
pattern_type=<pullback|breakout|reversal|trend_following|range_trading>,
market_condition=<trending_up|trending_down|ranging>,
timeframe=<derived from analysis>,
instrument_scope=[<equity|option|futures>]
)
- Filter results by
research_score(higher = more thoroughly researched). - For each matched setup, load confirmation signals and compare against current market data.
- Rank by fit: exact match → similar match → partial match.
- If no match: proceed with manual pattern rules from
references/chart-patterns-reference.md.
Step 5 — Pattern Output and Handoff
Produce structured pattern identification output:
- Primary pattern: The best-fit identified pattern with confidence level.
- Secondary patterns: Alternative patterns if primary is borderline.
- Invalid patterns: Patterns that appear but are invalidated by context (e.g., bearish engulfing in strong bull).
- Handoff to
trade-execution: Pattern type + market condition + timeframe → for setup selection and trade plan generation.
Output File
{ticker}/{YYYY-MM-DD_HHMM}_chart-patterns_{affix}.md
Output Structure
- Executive Summary — Primary pattern identified, market context, confidence level
- Context Assessment — Market structure (from price-action), Always In direction, valid/invalid pattern directions
- Trend Analysis — Bar counting (H1/H2/H3/H4 status), channel classification (tight/broad), breakout quality
- Reversal Assessment — 5-step sequence status (which steps have completed), climax detection, wedge count
- Pattern Details — Primary pattern with specific entry/stop criteria, secondary patterns
- Setup Matches — Results from
search_technical_setupsranked by fit; match quality assessment - Handoff Parameters — pattern_type + market_condition + timeframe for trade-execution skill
- Coverage Gaps — Data limitations and degraded-mode annotations
Error Handling
| Error | Fallback |
|---|---|
| No upstream context from price-action | Self-derive market structure from raw price data; flag degraded |
search_technical_setups unreachable | Provide manual pattern rules from references/chart-patterns-reference.md; flag gap |
| Pattern ambiguous | List all viable patterns; recommend waiting for confirmation bar |
| No matching pattern found | Report "no high-confidence pattern identified"; flag for manual review |
Memory Load
See contracts/memory-load.md.
Snapshot
See contracts/snapshot-synthesis.md.
Final Summary (TUI)
Include ### Key Citations block with 0-10 clickable /v/ URLs.
References
references/chart-patterns-reference.mdcontracts/citation-and-memory.mdcontracts/output-frontmatter-schema.mdcontracts/memory-load.mdcontracts/snapshot-synthesis.mdcontracts/preflight.mdcontracts/retrieval.md
Signals
- GitHub stars
- 204
- Forks
- 16
- Last commit
- Sep 2026
Advanced
- Catalog kind
- skill
- Gateway key
chart-patterns- Source
- github.com/agentii-ai/agentii-investment-intelligence