agentleFS
krzysukrzysu/market-skills/skills/strategy-funding-carry/SKILL.md

strategy-funding-carry

Skill3 starsChanged 4 months agoMITView on GitHub

L3 non-directional strategy. Harvests perpetual swap funding rates by entering the side that receives funding. Composes market-basis funding data.

The file

Files
SKILL.md

strategy-funding-carry

L3 non-directional strategy that harvests perpetual swap funding rates for income. Enters the side receiving funding (long when funding is negative, short when funding is positive) and exits when funding normalizes.

When NOT to use

  • Without a mandatory risk-engine vet before execution — this skill emits ideas, not orders. Always vet the Intent and get explicit user approval.
  • On spot markets — funding rates are a perpetual-swap construct. The strategy needs fetch_funding_rate to return a live rate, which requires a perp ticker (e.g. BTC/USDT on a CCXT exchange).
  • For a directional view — this is a carry trade, not a price-direction call. The entry direction is determined by which side receives funding, not by trend/momentum.
  • Without confirming the venue's funding mechanism — funding intervals, settlement, and sign conventions vary by exchange. The strategy reads the raw per-8h rate from market-basis/fetch_funding_rate.

Quick Start

uv run skills/strategy-funding-carry/scripts/run.py BTC/USDT --source ccxt:binance
uv run skills/strategy-funding-carry/scripts/run.py BTC/USDT --source ccxt:binance --json

Flags

FlagDefaultNotes
TICKER (positional)—Required. Supports provider:ticker (e.g. hl:LIT, yf:AAPL). Use a perp ticker for funding data.
--jsonhumanEmit JSON to stdout.
--source=PROVIDERauto-detectForce a data provider (see README). Funding rates require ccxt or ccxt:exchange.
--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.

Composes

L2 SkillPurpose
market-basisFunding rate data via fetch_funding_rate (raw per-8h rate, percentile mapping)
market-volatilityATR(14) for stop placement (L1, composed by L3 directly)

Entry Logic

  • Direction: negative perp funding (shorts pay longs) → LONG; positive perp funding (longs pay shorts) → SHORT. The carry trade enters the side that receives funding.
  • Funding extremity → conviction (absolute per-8h rate thresholds):
    • |rate| >= 0.001 (0.1% per 8h) → conviction 4
    • |rate| >= 0.0005 (0.05% per 8h) → conviction 3
    • |rate| >= 0.0001 (0.01% per 8h) → conviction 2
    • |rate| < 0.0001 → neutral, no idea
  • Entry: current price (last close)
  • Stop: 2 × ATR(14) from entry (long: entry − 2×ATR; short: entry + 2×ATR)
  • Targets: 3-TP ladder at 1.5R / 2.5R / 4R, with TP3 clamped at the 5% dead-zone boundary (entry × 1.05 long, entry × 0.95 short) so low-vol assets still emit an idea

Output

  • ideas[] — trade ideas with direction, conviction, entry/stop/target, reasoning
    • Each idea carries version: "v1".."v5" derived from conviction via analysis.contracts.conviction_version()
    • Each idea carries take_profit_ideal (unrounded construction) and rr_to_tp: [rr_to_tp1, rr_to_tp2, rr_to_tp3] (precomputed R:R to each TP via analysis.contracts.compute_rr_to_tp()) so consumers can read a canonical R:R without reimplementing the direction-asymmetric formula
    • Each idea is validated against validate_l3_tp_ladder() (TP3 ≥ entry × 1.05 long, or ≤ entry × 0.95 short). If validation fails after the clamp, the strategy surfaces the validator's error message as narrative so the silent-failure fingerprint doesn't reappear.
  • narrative — summary for user briefing (funding regime, direction, conviction)

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.