菜单
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.
用 Codex 或 Claude 帮你安装 复制这段 Prompt，粘贴到 Codex、Claude 或其他助手里，让它检查 Skill 页面并帮你完成安装。
基于 SOC 职业分类
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".
| name | sf-market-fit |
| description | 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. |
"Best strategy of the year" and "best strategy this week" are not the same thing. This skill picks strategies that are winning in the current market, not strategies with a fat 90-day backtest that has since gone cold.
For every candidate, compute annualized returns over three windows:
ann_7 = pct7Days × (365/7)
ann_30 = pct30Days × (365/30)
ann_90 = pct90Days × (365/90)
Then label:
accelerating — ann_7 > ann_30 > ann_90 and 7d positive (best fit)steady — recent windows positive and not collapsingfading — 7d down sharply vs 30d (recent regime hostile)stale — 30d non-positive or PF below 1.1 (not a current edge)Only accelerating and steady make the cut. fading is rejected here
but surfaced by sf-edge-watch.
python sf-market-fit/fit.py \
[--top N] # default 10
[--prescreen N] # default 150
[--min-pf X] # default 1.2
[--min-trades N] # default 30
[--pair CSV] [--timeframe CSV]
[--include-fading] # surface fading/stale too, with labels
[--json]
Ranked table with score, label (accelerating/steady/fading/stale), and the 3 recency windows so the user can see what changed.