Skip to content
SkillScript-led2,344 tokens in use

FTD Detector

Detects Follow-Through Day signals on the S&P 500 and Nasdaq Composite with a rally-attempt state machine and a 0-100 technical quality score.

7,427 stars on the repository MIT Licensetradermonty/claude-trading-skills · skills/ftd-detector

Skill label

measured from its files

What installing FTD Detector costs

111tokens idle
2,344tokens in use

Idle is what sits in context while the skill is installed. In use is its instructions, loaded when it runs. Counts are approximate, made offline with cl100k_base.

Model dependence Script-led

Script-ledMixedModel-led

Most of this skill is scripts, so it behaves much the same whichever model runs it.

This describes the skill's structure, how much of it is scripts. It does not say how well it works. Scripts: scripts/_price_bars.py, scripts/fmp_client.py, scripts/ftd_detector.py, scripts/ftd_lifecycle.py, scripts/post_ftd_monitor.py, scripts/rally_tracker.py, scripts/report_generator.py.

Touches

shelldetectednetworkdetectedwrites filesdetectedcredentialsdetected

Declared tags come from the skill's frontmatter; detected ones from its scripts, its code blocks and the commands it names in inline code. This is detected, not audited.

How it scores

Quality

How well the skill is written, counted from its own files: its frontmatter, its steps, examples, stated output and constraints, its size and whether every file it ships is one it names.How we score it →
Excellent
  • Frontmatter12/15
  • Structure23/25
  • Examples15/15
  • Output15/15
  • Size15/15
  • Resources1/15

Security

Starts at 100 and loses points for each thing in the skill's files that a reader should look at first.How we score it →
Good
  • Reads credentials−15

    does this on the machine, and the frontmatter does not say so

    SKILL.md, line 62
  • Runs shell commands−5

    does this on the machine, and the frontmatter does not say so

    SKILL.md, line 62
  • Reaches the network−10

    does this on the machine, and the frontmatter does not say so

    scripts/fmp_client.py, line 32
  • Writes files−8

    does this on the machine, and the frontmatter does not say so

    scripts/report_generator.py, line 17

Repository health

Pre-computed score based on development activity, maintenance, community, maturity, and trend momentum.How we score it →
Good
  • Development activity100
  • Maintenance52
  • Community68
  • Maturity36
  • Momentum40

FTD Detector runs a Python script that pulls S&P 500 and Nasdaq Composite history from the Financial Modeling Prep API and tracks a market correction through rally attempt, Follow-Through Day window and confirmation, using William O’Neil’s methodology.

It reports the current state, a 0-100 quality score for the signal, key watch levels and post-signal health such as distribution days. The score is a technical heuristic. The skill does not assess portfolio exposure and gives no investment advice.

What You Get

  • A JSON and a Markdown report named with the run timestamp
  • The combined market state across both indices, from no signal to confirmed
  • A 0-100 quality score with a signal label, marked as a technical heuristic
  • Key watch levels such as the swing low and the Follow-Through Day low
  • Post-signal health: distribution day count, invalidation check and power trend
  • Explicit data-gap flags when price bars are missing or invalid

Common Use Cases

  • Checking whether a bounce after a correction has produced a confirmed Follow-Through Day
  • Counting rally-attempt days and seeing which day the qualifying window is on
  • Reviewing the health of a signal after it appears, including distribution days
  • Pairing with a top-detection skill to look at both sides of a market cycle

What Makes It Unique

The skill keeps the event separate from its current health: a close below the signal-day low is a warning, while an intraday low below the recorded Day 1 low fails the rally. Missing or invalid bars are never read as zero; they produce an explicit data-gap state and a score of zero. The exposure field always says Not assessed, so the score is never turned into a portfolio percentage.

Under The Hood

Claude runs a Python script (ftd_detector.py) from the skill folder with a Financial Modeling Prep API key from the FMP_API_KEY environment variable or a command-line flag. The script makes about four calls to the FMP web API for index history and quotes, runs the state machine locally and writes the JSON and Markdown reports to disk. Reference documents on the method and post-signal health are loaded only when needed. Python 3.9 or later with the requests library is required.

Join founders building with open source

Opinionated takes, migration guides, cost-saving tips, and insights from the open source ecosystem.

Subscribe on SubstackJoin 750+ subscribers