claude-skills

Pick the best StrategyFactory strategies using a multi-factor composite score (profit factor + sharpe + drawdown + recency + sample size). Use whenever the user asks "what's the best strategy on SF / pick me the top N / find me bangers / which strategies should I deploy". Returns a ranked shortlist with per-factor breakdown so the agent can explain WHY each pick made it. Read-only.

Deploy a StrategyFactory strategy as a copy bot on one of your connected exchanges. Use only after the user has explicitly chosen a strategy (typically via sf-best-pick / sf-market-fit / sf-profit-factor) and named the target exchange and sizing. Dangerous — creates a live bot that places real orders. Always confirm with the user before invoking.

Diagnose whether a strategy's edge is alive, softening, decaying, or dead — by comparing forward-test vs backtest profit factor / win rate and 7d vs 30d vs 90d recency. Use when the user asks "is this strategy still working / is the edge alive / is X dying / health-check this strategy". Read-only.

List HOT StrategyFactory strategies with optional quant filters and sorting. Use when the user asks "what's hot on StrategyFactory", "scan SF for live strategies", "show me HOT strategies with PF > 1.5", or any variant of "what's on the SF leaderboard". Read-only, requires SF_API_KEY/SF_API_SECRET.

Recurring scanner that watches the StrategyFactory HOT leaderboard and surfaces NEW top-N picks as they appear. Designed to be wrapped by the /loop skill (or any cron) — runs one pass per invocation, persists the previous top-N to disk, and emits a delta. Use when the user asks "keep an eye on SF for new winners / monitor the leaderboard / alert me when a new top edge shows up / find me new bangers automatically".

Recurring health watch for deployed bots. One invocation = one pass. Flags decaying / dead bots, can auto-pause them, persists a log so consecutive runs don't double-fire. Pair with the /loop skill for an always-on roster cleaner. Use when the user asks "watch my bots for edge decay automatically / auto-pause dying bots / kill dead edges on a loop / monitor roster health".

Find StrategyFactory strategies that are winning RIGHT NOW in the current market regime — accelerating 7-day returns, strong 30-day, with edge still intact. Use when the user asks "what's working this week / which strategies fit the current market / find strategies that match BTC's regime / show me what's hot the last 7 days". Read-only.

One-screen account view across StrategyFactory — exchanges, cached USDT balances, bots (with status), and open positions per exchange. Use when the user asks "where do I stand / show my portfolio / what bots are running / what's my balance / what positions are open / portfolio snapshot". Read-only.

Rank StrategyFactory strategies specifically by profit factor with sanity floors so PF outliers from tiny samples don't dominate. Use when the user asks "find high profit factor strategies / show me PF > 2 / which strategies have the best risk-reward / what's got the cleanest profit factor". Read-only.

Find deployed StrategyFactory bots whose underlying strategy's edge is decaying, and optionally pause them. Use when the user asks "kill the dead bots / prune my portfolio / pause the losers / which of my bots are dying / clean up the bot roster". Writes state via PATCH only when --apply is passed.

Refresh the deployed bot roster — pause decaying bots AND propose fresh top picks to replace them on the same exchanges with the same sizing. Use when the user asks "rebalance my bots / swap losers for winners / refresh the roster / clean up and replace / optimize my deployed strategies". Writes state (pauses) only with --apply; never auto-deploys.

Scan Bybit USDT perpetuals for short-term volatility spikes. Use when the user asks for scalp candidates, momentum plays, "what's moving on Bybit", or any variant of "find me a trade / find movers / scan perps". Read-only, no API keys required.

Summarize short-term price structure for a single Bybit USDT perpetual — trend vs range, recent swing high/low, VWAP, ATR. Use after bybit-scan picks a candidate, or whenever the user asks "what's the structure on X / is X trending / where are the levels on X". Read-only, no API keys.

Place a market entry with a native bracket stop-loss and take-profit on Bybit v5. Testnet by default. Use ONLY when the user explicitly asks to place/execute/open a trade and has already planned size with scalp-plan and confirmed risk with risk-guard. Dangerous — authenticated, touches real money on --live. Always confirm with the user before invoking.

Single source of truth for trading session P&L, per-trade risk cap, and daily drawdown kill-switch. Always called by bybit-trade before any order; also exposes status/init/open/close subcommands. Use for "what's my P&L / am I within risk limits / reset my session". Stateful, no network, no API keys.

Compute position size, stop, take-profit, notional and margin for a scalp trade given equity, $-risk, entry, and stop/target percentages. Use whenever the user asks "how much size / size this trade / plan this scalp / what's my stop and target". Pure math, no network, no API keys.