trader-scanner

The Chart Scanner is the first step of a day trade. It reads thousands of free charts and lists the ones that match a rule. That is all it does. A scan is a list of charts that match a rule, not a list of stocks to buy, and a setup on a chart is not a prediction. Most people who day trade lose money; the scanner does not change that. It only saves you from flipping through 5,000 charts by hand.

${CLAUDE_SKILL_DIR} below is this skill's folder (where this SKILL.md lives).

python3 "${CLAUDE_SKILL_DIR}/scan.py" universe                       # -> trader/universe.txt
python3 "${CLAUDE_SKILL_DIR}/scan.py" daily                          # before the open -> trader/watchlist.json
python3 "${CLAUDE_SKILL_DIR}/scan.py" daily --symbols TSLA,NVDA,AMD  # or --watchlist FILE, --limit 200
python3 "${CLAUDE_SKILL_DIR}/scan.py" orb                            # after 9:50 AM ET -> trader/watchlist.json
python3 "${CLAUDE_SKILL_DIR}/scan.py" orb --at 10:15                 # cut the clock at 10:15 AM ET
python3 "${CLAUDE_SKILL_DIR}/scan.py" orb --watchlist trader/watchlist.json --json

Flags: --min-price 5 and --min-volume (daily 1,000,000, orb 0), --or-minutes 15, --max-confirm 200 (orb stage 2), --now "2026-10-19 10:15" (New York time) to pin the clock, --offline PATH to read saved Yahoo files, --no-write, --show N, --json. Exit code 0 when the scan found something, 1 when it found nothing or could not run (the reason is one plain line).

What it reads and writes

file	what
trader/universe.txt	written by universe: one SYMBOL|Name a line, read by daily and orb
trader/watchlist.json	written by daily and orb: mode, as_of, universe, scanned, failed, hits[]. trader-news and trader-entry read hits[].symbol
trader/config.json	read by orb: no_entries_after (default 15:30)

The universe

universe downloads nasdaqlisted.txt and otherlisted.txt from nasdaqtrader.com/dynamic/SymDir and keeps a row when, in this order:

1. It is not a test issue, not an ETF, not NextShares.
2. It is a plain ticker: 1 to 5 letters, or 1 to 4 letters plus one class letter after a dot (BRK.B). A $ preferred (ABR$D) is out, and so is a .U, .W or .R suffix (units, warrants, rights).
3. Out, whatever else the name says: warrants; SPAC units and rights ("Units, each consisting of", "Rights to receive"); notes, ETNs, debentures; a % coupon; "Depositary Shares" that are not American or Global (those carry a preferred).
4. In: the name calls it common stock, common shares or units, ordinary shares, shares of beneficial interest, ADS / ADR / American or Global depositary shares, or partnership or trust units.
5. Out: the name still says unit, right, bond, preferred or preference.
6. In: everything else (plain names like "AMETEK, Inc.").

On the Oct 19 2026 files this keeps 5,943. The reel said 5,259: that count came from an ad hoc filter (names that say common or ordinary, no ADRs). All 5,259 of those are in this list; the 684 extra are mostly ADRs, closed-end funds and partnership units. The Oct 20 files gave 5,942.

The daily scan (before the open)

Yahoo's spark endpoint returns closes for 20 symbols a request, so 5,942 symbols take 298 requests. On Oct 20 2026 that took 145 seconds with the polite pause (0.3 s). Liquidity comes from the spark meta: last price at least --min-price and last session's volume at least --min-volume. Run it before the open or after the close: during the session the volume is the day's so far, and today's daily bar is left out until 4:00 PM.

The setups are computed from daily closes only (spark has no highs or lows), so they are close cousins of the reel's refs/scan.py, which used highs and lows:

setup	the rule on closes
breakout_20d	the last close is above every close of the 20 sessions before it
bull_flag	up more than 10 % over the 10 sessions before the last 5, then the last 5 closes inside a 5 % range
reclaim_ma20	the close before was under its 20-day average, the last close is above it
oversold	RSI 14 (Wilder) under 30
oversold_bounce	RSI 14 was under 30 yesterday and today closed up

A closing-high breakout fires more often than an intraday-high one (KO on Oct 16 broke its 20-day closing high, not its 20-day intraday high). Hits are listed most traded first (price times volume).

The opening-range breakout scan (during the session)

The opening range is the first --or-minutes (15) of the session: the 9:30, 9:35 and 9:40 five-minute bars. A hit is the first completed 5-minute bar to CLOSE above the range high, at that bar's close time.

1. Stage 1, every symbol: spark closes of today's 5-minute bars, 20 a request. A chart goes on only if a completed bar after the range closed at or above the highest close of the range. That is a necessary condition (the range high is at least every range close), so no breakout is lost.
2. Stage 2, each candidate: the v8 chart with real highs and lows, one request each. Range high = the highest high of the range bars, range low = the lowest low. On a strong morning most charts are candidates, so stage 2 checks at most --max-confirm (200) of them, the most traded first (price times volume so far), and prints PARTIAL with the count it skipped. --max-confirm 0 checks them all.

Timing, live on Tue Oct 20 2026 (a strong morning): 100 liquid names at 10:01 AM took 35 seconds (5 spark requests, 70 v8 charts, 51 of 91 usable charts broke out). The whole universe at 10:03 AM took about 4 minutes (298 spark requests, then 200 v8 charts): 2,448 usable charts, 1,782 candidates, the 200 most traded checked, 119 of those broke out. For a fast answer, scan the daily watchlist or a list of names you trade (--watchlist, --symbols).

A chart is usable only when every 5-minute bar from 9:30 to the cutoff is there; a missing bar means a halt or a thin stock. Null closes are dropped, a bar off the 5-minute grid (the live tick, like 10:58:07) is dropped, and a bar counts only once it is complete (start + 5 minutes). Before 9:50 AM, on a weekend or holiday, after the close, or after no_entries_after (an expired setup) it prints one plain line and writes nothing. The 1:00 PM early closes are in the calendar.

Data and limits

• Yahoo's chart and spark endpoints are free, unofficial and can lag. Bars are 5-minute bars, bar by bar, never tick by tick. query2 first, query1 on a 429 or 5xx, one retry after a 5-second wait on a 429.
• Partial results are said as partial: "Fetched 4,980 of 5,259 charts (279 failed or had no data)". Never round a partial scan up to "the whole market".
• NYSE holidays and early closes for 2026 and 2027 (nyse.com, read 2026-10-13).

Steps

1. Once a week or so: scan.py universe.
2. Before the open: scan.py daily. Tell the user the counts (fetched, liquid, scanned, hits) and the top few by setup.
3. After 9:50 AM ET (10:15 is the classic): scan.py orb. Tell the user how many usable charts broke their range and when, and that a later hit is closer to the price now.
4. The user picks one chart. Hand it on: /trader-news TICKER (what is moving it), then /trader-entry TICKER (the entry bar by bar). Never pick for them.

Example: the reel

Monday Oct 19 2026, scan at 10:15 AM over the reel's 234-ticker wall (tests/fixtures/trader-scanner/wall5m-2026-10-19.json.gz, real Yahoo charts):

python3 scan.py orb --offline wall5m-2026-10-19.json.gz --now "2026-10-19 10:15" --min-price 0
Fetched 233 of 234 charts (1 failed or had no data); 231 usable; 95 candidates, 95 confirmed with real highs and lows.
42 of 231 charts closed above their opening-range high by 10:15 AM:
   9:50 AM  AMC         $2.97  range 2.90 to 2.96  +0.50% over
  ...
  10:15 AM  TSLA      $364.18  range 354.47 to 362.18  +0.55% over

The reel's wall had no price floor (AMC at $2.97 is one of the 42), hence --min-price 0; with the default $5 floor it is 41. TSLA's range is the 9:30 to 9:40 bars, high 362.18, low 354.47; the 10:10 bar closed at 364.18, 0.55 % over the high, at 10:15 AM. The user picks TSLA and Claude hands it to /trader-news TSLA, then /trader-entry TSLA. In the reel the universe was 5,259 US stocks; this tool's rule gives 5,943 on the same files.

The pre-market scan that morning (closes through Fri Oct 16, the reel's daily charts) tags BLK, CI, XOM and HD a 20-day breakout, CCL a bull flag, F, JPM and GE a reclaim of the 20-day average, and BAC, MS, RIVN and PLUG oversold, as the reel did. On closes it also tags KO (breakout), RCL (flag) and PNC (an oversold bounce).

Rules

• Say what was scanned: how many symbols, how many charts came back, how many were usable, and the time of the data.
• Say "matched a rule", never "a buy", "a winner" or "a good trade".
• Never print a P&L, a return or a profit figure.
• A hit is stale once the price has moved on. Re-run before acting on it.

What it never does

It never places, sizes or checks an order, never says which stock to buy, never calls data real time, and never promises that a setup works.

Safety

• Education and tooling, not financial advice. A PASS is a rules check, not a recommendation.
• Paper trading by default ("paper": true). Going live is the user's own decision, in their own config.
• Nothing in this pack places an order except trader-broker, and only when all three hold: trader-risk passed that exact order today, the user typed an explicit yes for that one order in the chat, and the user clicked Allow once on Claude's own card because the order tool is set to "Needs approval" (Customize > Connectors > tool permissions). Never set an order tool to "Always allow".
• No options, margin, short selling or leverage unless the config allows it and the user confirms.
• Broker keys live in environment variables or the Claude Desktop config, never in the chat.