agentleFS
krzysukrzysu/market-skills/skills/market-breakout/SKILL.md

market-breakout

Skill3 starsChanged 4 months agoMITView on GitHub

Detects whether a breakout is real by composing L1 indicators: structure break, volume confirmation, OBV confirmation, squeeze release, and retest holding. Classifications: FRESH, CONFIRMED, STALE, FAILED.

The file

We split this skill into the rules it gives an agent

2 rules. About 0 are common, 0 are distinctive, and 2 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈0

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 2

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Very specific:

    Do not treat STALE or FAILED classifications as reversal entries.

    line 20
  2. Very specific:

    When the primary trigger isn't met (specifically: fewer than 2 sub-signals present, or weighted_sum <= 0.30) but a known combination of …

    line 65

≈ “Common” is an estimate: at least 1 in 5 of the 372,429 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

market-breakout

L2 pattern detection skill that composes L1 indicators to determine whether a breakout is genuine or likely to fail.

When NOT to use

  • This is pattern detection, not a trade idea — feed a fired pattern into strategy-breakout-confirm or run-all-l3 before acting.
  • In low-volume or choppy tape a FRESH/CONFIRMED breakout is a frequent fakeout — require volume confirmation (market-volume) and squeeze release (market-squeeze) before trusting it.
  • Do not treat STALE or FAILED classifications as reversal entries; they only say the breakout did not hold.

Flags

FlagDefaultNotes
TICKER (positional)—Required. Supports provider:ticker (e.g. hl:LIT, yf:AAPL).
--jsonhumanEmit JSON to stdout.
--source=PROVIDERauto-detectForce a data provider (see README).
--interval=INTERVAL1d1m/2m/5m/15m/30m/1h/2h/4h/8h/12h/1d/3d/1wk/1M.
--period=PERIOD1y1d/5d/1mo/3mo/6mo/1y/2y/5y/10y/ytd/max.

Both timeframe flags are validated — bad values exit 2 with a friendly error. For intraday (--interval=1h), bump --period to 6mo or 1y; yfinance caps hourly at ~2y and anything sub-hour at ~60d.

Classifications

ClassificationMeaning
FRESHBreakout just initiated (confidence >= 3, no retest yet)
CONFIRMEDBreakout confirmed by successful retest of the breakout level
STALEBreakout has been present >10 bars without continuation
FAILEDStructure break reversed, price returned to consolidation

Sub-signals

Sub-signalWeightSource L1Logic
Structure break0.35market-trendalignment is FULL_BULL or FULL_BEAR, or signal is STRONG_UPTREND/STRONG_DOWNTREND
Volume confirmation0.25market-volumevolume_ratio > 1.5
OBV confirmation0.15market-volumeobv_trend == "rising" with bullish, or "falling" with bearish
Squeeze release0.15market-squeezesqueeze_on is False (released) and direction == "increasing" with bullish momentum
Retest holding0.10market-s-rsits_on_level is True after break (lagging)

Trigger

pattern.present is True when at least 2 sub-signals are present AND their combined weight exceeds 0.30. The previous weighted_sum / total_weight >= 0.40 threshold silently dropped 2-sub combinations at weighted_sum in (0.30, 0.40) — e.g. volume_confirmation + retest_holding at 0.35 — into present=False while the sub-signals were populated.

confidence is round(weighted_sum * 5), clamped to [1, 5].

Recognized sub-shapes

When the primary trigger isn't met (specifically: fewer than 2 sub-signals present, or weighted_sum <= 0.30) but a known combination of sub-signals is firing, the classifier falls back to a recognized sub-shape. This catches 2-sub combos below the primary trigger's 0.30 threshold — e.g. squeeze_release + retest_holding (wsum 0.25) where two corroborating L1s (squeeze + S/R) plus the implied direction from squeeze are meaningful.

Sub-shapeSub-signals requiredClassification
Post-squeeze retest holdingsqueeze_release + retest_holdingCONFIRMED

Output

  • pattern.present (bool)
  • pattern.confidence (1–5)
  • pattern.classification (FRESH / CONFIRMED / STALE / FAILED)
  • pattern.type always "BREAKOUT"
  • signals — per-signal {"present": bool, "weight": float}
  • input_scores — raw L1 outputs
  • narrative — one-sentence explanation

Output envelope (AXI)

--json output follows the canonical AXI envelope — {data, count, errors, help[]}. Default schema is the per-skill minimal fields (3-6 essentials); pass --fields=<csv> to project or --full for the full payload. count is the item count, help[] is contextual next-step command templates. Lib.py return shapes (L1Result / L2Result / L3Result / L3Idea / RegimeSignal) are unchanged — the envelope wraps them at the scripts/run.py boundary.

Home view (no-arg mode)

No-arg mode prints the home view from $XDG_DATA_HOME/market-skills/<skill>_last.json (last successful run). Errors are not cached.

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.