Index of every page on this site: llms.txt
All skills
tradermonty avatar

/ibd-distribution-day-monitor

@f95e770
Run it yourself

Detect IBD-style Distribution Days for QQQ/SPY (close down at least 0.2% on higher volume), track 25-session expiration and 5% invalidation, count d5/d15/d25 clusters, classify market risk (NORMAL/CAUTION/HIGH/SEVERE), and emit TQQQ/QQQ exposure recommendations. Use after market close, before TQQQ exposure changes, or as input to FTD/market-state frameworks. Does not execute trades.

Use this Skill: https://skilld.dev/gh/tradermonty/claude-trading-skills/ibd-distribution-day-monitor

Nothing lands on disk. Nothing to clean up.

Fork this Skill

Edit a local copy. It keeps the author and licence.

SKILL.md

≈104 tokens for metadata: the name and description. ≈1.5k when used: this file. ≈2k more on demand in 3 files.

Description uses 5.2% of example budget

Before choosing a Skill, your Agent reads its name and description. All available Skills share that space.

  • A shorter description leaves more room for other Skills. This entry exceeds our 1% size suggestion.

In our Claude Code example, all Skill names and descriptions share 8,000 characters. This Skill uses ≈416 characters, or 5.2%.

The 1% threshold is a size suggestion. Longer descriptions can still fit.

Your model, settings, and other Skills decide how much text your Agent can read.

Example settings and source

The example uses a 200k-token context and default Claude Code settings. The count includes the name, description, separators, and when_to_use when present. Codex also counts local file paths.

Skit's source and limits: Codex 0.160.1, Claude Code 2.1.292.

IBD Distribution Day Monitor

Instrument identity (#765)

QQQ is a Nasdaq-100 ETF proxy, not the Nasdaq Composite (^IXIC). SPY is an ETF proxy, not ^GSPC. DD counts use the named symbol's own price and volume; ETF counts are not official index counts. Select explicit symbols with --symbols; missing data must not trigger substitution or a mixed series. The risk classes and TQQQ policy are this tool's separate heuristics, not a calibrated FTD success probability (calibration remains #676).

Purpose

Detect IBD-style Distribution Days for major market ETFs (QQQ as Nasdaq proxy, SPY as S&P 500 proxy) and produce a daily market deterioration signal plus a TQQQ/QQQ exposure recommendation. Designed for post-market review.

When to Use

Invoke this skill:

  • Daily after the US market close.
  • Before increasing TQQQ exposure or rebalancing leveraged positions.
  • When evaluating whether an uptrend is becoming vulnerable to a correction.
  • As an upstream input to FTD (Follow-Through Day) detection or other market-state frameworks.

Do NOT use this skill to:

  • Execute trades or modify orders.
  • Generate discretionary market predictions outside of the IBD ruleset.

Prerequisites

  • FMP API key: Required. Pass --api-key, set config.data.api_key, or set the FMP_API_KEY environment variable, in that priority order.
  • Other API keys: None used.
  • Python: 3.9+ with PyYAML>=6.0 and requests>=2.31.0 (see requirements.txt).

Inputs

  • Symbols (default: QQQ, SPY) and lookback (default 80 trading sessions).
  • Optional --as-of YYYY-MM-DD for backtesting against a historical session.
  • Optional --config for a readable UTF-8 YAML mapping. Empty/null documents use defaults. Configuration read/parse errors and invalid top-level, data, strategy_context or indexes shapes return exit 1 on stderr before provider access; correct the file/path and retry. YAML error diagnostics do not echo file contents.
  • Strategy context: instrument (TQQQ or QQQ), current exposure %, base trailing stop %.
  • FMP API key via --api-key, config.data.api_key, or FMP_API_KEY env var (in that priority order).

Core Rules

A Distribution Day is detected when:

  1. Today's close is at least 0.2% below yesterday's close.
  2. Today's volume is greater than yesterday's volume.

A Distribution Day is removed from the active count when either:

  • More than 25 trading sessions have elapsed since the DD.
  • The index has gained 5% from the DD close (using post-DD high by default; configurable to close-source).

Today's DD is never invalidated immediately because there are no post-DD sessions to evaluate the 5% gain against.

Counting Conventions

  • d5_count / d15_count / d25_count count active records with age_sessions <= N.
  • This means N+1 sessions are inspected (age 0..N inclusive). Reports therefore say "within N elapsed sessions" rather than "直近 N 取引日" to avoid ambiguity.

Risk Classification

Risk Trigger
NORMAL d25 <= 2
CAUTION d25 >= 3
HIGH d25 >= 5 OR d15 >= 3 OR d5 >= 2
SEVERE d25 >= 6 OR d15 >= 4 OR (market_below_21ema_or_50ma AND d25 >= 5)

When both QQQ and SPY are loaded, QQQ-weighted overall logic applies (TQQQ-aware): a single SEVERE escalates to SEVERE; QQQ HIGH escalates to overall HIGH; QQQ NORMAL + SPY HIGH still escalates to HIGH (broad-market spillover).

TQQQ Exposure Policy

Risk Action Target Exposure Trailing Stop
NORMAL HOLD_OR_FOLLOW_BASE_STRATEGY 100% base
CAUTION AVOID_NEW_ADDS 75% min(base, 7%)
HIGH REDUCE_EXPOSURE 50% min(base, 5%)
SEVERE CLOSE_TQQQ_OR_HEDGE 25% min(base, 3%)

QQQ uses a less aggressive policy (HIGH=75%, SEVERE=50%) since it lacks 3x leverage.

Workflow

  1. Load OHLCV for the configured symbols via FMP (get_historical_prices).
  2. Validate data quality; record skipped sessions in audit.
  3. Rebase via prepare_effective_history so effective_history[0] is the evaluation session.
  4. Detect raw Distribution Days; enrich with high_since, invalidation event, and status.
  5. Count d5 / d15 / d25 active records.
  6. Compute 21EMA and 50SMA filters; flag market_below_21ema_or_50ma (None if data insufficient).
  7. Classify each index, then combine using QQQ-weighted policy.
  8. Generate portfolio action for the configured instrument.
  9. Write JSON + Markdown reports to --output-dir with API keys redacted.

Outputs

Saved to reports/ (or --output-dir):

  • ibd_distribution_day_monitor_YYYY-MM-DD_HHMMSS.json
  • ibd_distribution_day_monitor_YYYY-MM-DD_HHMMSS.md

JSON is UTF-8 with ensure_ascii=False (Japanese explanations preserved). Sensitive keys (api_key, fmp_api_key, token, etc.) are redacted automatically.

Operating Principles

  • Do not override the IBD rule definitions unless config/default.yaml is changed deliberately.
  • Always explain which dates contributed to the active count.
  • Treat missing or unreliable volume data as a warning (audit_flag), not as a Distribution Day.
  • Do not place trades. The portfolio action is a risk-management suggestion, not an execution instruction.

CLI

python3 skills/ibd-distribution-day-monitor/scripts/ibd_monitor.py \
  --symbols QQQ,SPY \
  --lookback-days 80 \
  --instrument TQQQ \
  --current-exposure 100 \
  --base-trailing-stop 10 \
  --output-dir reports/

API Requirements

FMP API key required. Free tier (250 calls/day) is sufficient for daily QQQ + SPY runs.

Related Skills

  • ftd-detector: Bottom confirmation via Follow-Through Days (counterpart of this top-side signal).
  • market-top-detector: Composite 0-100 top probability score using O'Neil distribution + other components.
  • position-sizer: Convert risk-management recommendations into share counts.

Source: SKILL.md on GitHub

skilld matched fixed text patterns in SKILL.md and file names. Patterns miss obfuscated code.

skilld run checks every file with the same patterns. It asks for approval before it loads a Skill with a behavior marked Needs approval.

No alerts3d3 checks · Risk SAFE
  • Gen Agent Trust Hub3d

    The skill is a market analysis tool that detects market distribution clusters and provides exposure recommendations for ETFs. It follows security best practices, including robust redaction of API keys to prevent credential leakage and the use of safe YAML loading for configuration. No malicious patterns or security risks were identified.

  • Socket3d

    No alerts

  • Snyk3d

    Risk: LOW · No issues

Signed by skilld at f95e770. This ties the file your Agent reads to that commit on GitHub. It does not review the instructions.

Last checked against GitHub yesterday.

Activeupdated 5 days ago

README badge

README badge for tradermonty/claude-trading-skills/ibd-distribution-day-monitor