Skills Plugins MCP Prompt Model 博客 我的中心
Education #trading #agent

trading-wisdom

Core trading insights learned from Agent Arena competition. Use when making any trading decision to apply institutional knowledge.

DeepseekModel Curated skill Quality Good · 64 v1.0.0

Get

https://deepseekmodel.com/api/download.php?id=0xhubed-agent-trading-arena-claude-skills-trading-wisdom-skill-md&format=skill
Download .skill Standard format with system_prompt and model_config, ready for any agent framework
The actual content of the system_prompt field in the .skill file.
name trading-wisdom description Core trading insights learned from Agent Arena competition. Use when making any trading decision to apply institutional knowledge. Trading Wisdom Last updated: 2026-03-09 20:08 UTC Active patterns: 232 Total samples: 22482 Confidence threshold: 60% Key Learnings Market was uniformly bullish (BNB +3.31%, BTC +2.68%, ETH +3.90%, SOL +5.07%, DOGE +2.46%) — the 7th consecutive window where regime misidentification was the primary loss driver. Only 1 of 6 active agents was profitable (journal_aware +$55.48). The other 5 active agents lost a combined -$571.18, suggesting widespread SHORT bias despite uniformly positive market. gptoss_skill_aware was the worst performer (-$340.51 on 31 trades), consistent with the persistent pattern of sophisticated validation frameworks providing false confidence on wrong-direction trades. Self-reflective position management (journal_aware) continues to be the most reliable edge among active agents, now profitable in multiple bullish windows. Zero trading (ta_bot, index_fund) outperformed 5 of 6 active agents, reinforcing that inaction beats wrong-direction action. SOL was the best performer (+5.07%) — highest-beta assets continue to show the largest moves, making them both the best LONG targets and the worst SHORT targets. Trade frequency amplifies losses when directional bias is wrong: skill_aware's 31 trades at -$10.98/trade vs contrarian's 6 trades at -$3.79/trade. Winning Strategies skill_aware_oss: High-frequency SHORT-biased tradi... Confidence : 95% Total samples : 200 Times confirmed : 1 First seen : 2026-02-01 Details : skill_aware_oss: High-frequency SHORT-biased trading (200 trades/24h) with multi-timeframe bearish alignment validation, 2% equity risk sizing, and disciplined profit-taking. Achieved +$2911.52 in uniformly bearish market (-6% to -11% across all assets). Combining technical validation ('risk calculator s... Confidence : 92% Total samples : 200 Times confirmed : 1 First seen : 2026-02-01 Details : Combining technical validation ('risk calculator shows 2:1 reward', 'validation permits trade') with trend alignment across 15m/1h/4h timeframes for SHORT entries in bearish markets. skill_aware_oss used this consistently with high confidence (0.78-0.92). index_fund's zero-trade strategy preserves capital... Confidence : 90% Total samples : 244 Times confirmed : 1 First seen : 2026-02-02 Details : index_fund's zero-trade strategy preserves capital ($0 PnL) when all active trading agents lose money. In uniform_bullish markets (+1.33% to +3.82% across all assets), passive allocation beats active SHORT-biased trading. Position sizing at 2% equity risk with 2:1 reward ... Confidence : 90% Total samples : 183 Times confirmed : 1 First seen : 2026-01-31 Details : Position sizing at 2% equity risk with 2:1 reward ratio SUCCEEDS when combined with validation checks AND market direction alignment. skill_aware_oss reasoning shows 'excellent 2:1 risk-reward with 2% equity risk' paired with bearish bias confirmation. Closing SHORT positions 'to lock in gains' when pr... Confidence : 88% Total samples : 172 Times confirmed : 1 First seen : 2026-01-31 Details : Closing SHORT positions 'to lock in gains' when profitable is correct behavior in bearish markets. skill_aware_oss: 'closing BNBUSDT locks in profit and reduces exposure' with 0.92 confidence. Closing losing LONG positions quickly ('close exis... Confidence : 88% Total samples : 50 Times confirmed : 1 First seen : 2026-02-01 Details : Closing losing LONG positions quickly ('close existing SOL long to free margin before initiating ETH short') to pivot to trend-aligned SHORT positions. skill_aware_oss demonstrated this adaptive behavior. skill_aware_oss maintains disciplined SHORT-biased... Confidence : 87% Total samples : 183 Times confirmed : 1 First seen : 2026-01-31 Details : skill_aware_oss maintains disciplined SHORT-biased strategy in uniformly bearish markets (-7.95% to -16.34% across all assets), using multi-timeframe bearish confirmation with selective position management. Closes positions to 'lock in profit and reduce exposure' rather than panic-cutting small losses. skill_aware_oss achieves highest PnL (+$409.48) wi... Confidence : 85% Total samples : 172 Times confirmed : 1 First seen : 2026-01-31 Details : skill_aware_oss achieves highest PnL (+$409.48) with high trade frequency (172 trades/24h) using SHORT-biased strategy in mixed-bullish market. Key: multi-timeframe bearish alignment (15m, 1h, 4h), 2:1 risk/reward ratio, and disciplined profit-taking on shorts when validation confirms. ta_baseline's low-frequency technical analysis app... Confidence : 85% Total samples : 38 Times confirmed : 1 First seen : 2026-02-02 Details : ta_baseline's low-frequency technical analysis approach (38 trades/24h) minimizes losses in bullish markets where SHORT-biased strategies fail. Selective trading with clear technical signals outperforms high-frequency approaches when market direction contradicts agent bias. Zero trading preserves capital perfectly when acti... Confidence : 85% Total samples : 2 Times confirmed : 1 First seen : 2026-03-09 Details : Zero trading preserves capital perfectly when active agents lose money. ta_bot (0 trades, $0) and index_fund (0 trades, $0) avoided all losses while 5 of 6 active agents lost money totaling -$571.18. Closing SHORT positions 'to lock in gains' when pr... Confidence : 83% Total samples : 172 Times confirmed : 1 First seen : 2026-01-31 Details : Closing SHORT positions 'to lock in gains' when profitable combined with re-entering on bearish confirmation signals. skill_aware_oss demonstrates this cycle repeatedly with BTC shorts. skill_aware_oss achieves highest PnL (+$691.44) wi... Confidence : 82% Total samples : 49 Times confirmed : 1 First seen : 2026-01-30 Details : skill_aware_oss achieves highest PnL (+$691.44) with moderate-high trade frequency (49 trades/24h) using SHORT positions in a uniformly bearish market. Key: Correctly identifying market direction and trading WITH the trend, not against it. Forum-aware selective trading with moderate freque... Confidence : 82% Total samples : 24 Times confirmed : 1 First seen : 2026-03-08 Details : Forum-aware selective trading with moderate frequency (24 trades from 89 decisions, 27% action rate) in a mixed/slightly-negative market. gptoss_forum_aware achieved +$232.17, the best performer by far, suggesting forum/social sentiment signals correctly filtered out low-conviction trades and identified profitable opportunities. Ultra-low frequency or zero trading preserves capi... Confidence : 80% Total samples : 89 Times confirmed : 1 First seen : 2026-03-08 Details : Ultra-low frequency or zero trading preserves capital when market direction is mixed/uncertain. index_fund (0 trades, $0 PnL) avoided all losses in a market where the momentum and contrarian agents lost money. ta_baseline achieves positive PnL (+$133.04) with ... Confidence : 80% Total samples : 36 Times confirmed : 1 First seen : 2026-01-31 Details : ta_baseline achieves positive PnL (+$133.04) with moderate trade frequency (36 trades/24h) in mixed-bullish market. Technical analysis baseline with selective trading outperforms aggressive LLM-based strategies. ta_baseline achieves positive PnL ($+20.54) with e... Confidence : 80% Total samples : 20 Times confirmed : 1 First seen : 2026-01-31 Details : ta_baseline achieves positive PnL ($+20.54) with extremely low trade frequency (20 trades vs 183 for top performer) by using simple SMA crossover signals and taking profit at defined ROE targets (6.1% ROE). Avoids overtrading in volatile bearish conditions. Zero trading or passive allocation preserves capit... Confidence : 80% Total samples : 12 Times confirmed : 1 First seen : 2026-03-07 Details : Zero trading or passive allocation preserves capital in bearish markets. index_fund ($0 PnL, 12 trades all passive allocations) and ta_bot ($0 PnL, 0 trades) avoided all losses while active agents with wrong bias lost $12-$108. In uniformly bearish markets (all assets -1.3% to ... Confidence : 78% Total samples : 97 Times confirmed : 1 First seen : 2026-01-30 Details : In uniformly bearish markets (all assets -1.3% to -2.3%), SHORT-biased strategies with RSI oversold confirmation and bearish MACD outperform. skill_aware_oss reasoning: 'RSI ~28, price below SMA, bearish MACD, multi-timeframe downtrend' for shorts. Journal-aware self-reflective position management ... Confidence : 78% Total samples : 27 Times confirmed : 1 First seen : 2026-03-09 Details : Journal-aware self-reflective position management with moderate trade frequency (27 trades) in a uniformly bullish market (+2.5% to +5.1%). gptoss_journal_aware achieved +$55.48, the only profitable active agent, likely by recognizing overtrading tendencies and closing positions to reduce exposure. Moderate trade frequency (48-63 trades/24h) outper... Confidence : 77% Total samples : 160 Times confirmed : 1 First seen : 2026-01-30 Details : Moderate trade frequency (48-63 trades/24h) outperforms both extremes in bearish markets. Too few trades (4-7) miss opportunities; the sweet spot captures trend moves without overtrading. Forum-aware selective trading with moderate freque... Confidence : 77% Total samples : 31 Times confirmed : 1 First seen : 2026-03-07 Details : Forum-aware selective trading with moderate frequency (31 trades) in a uniformly bearish market. gptoss_forum_aware achieved +$250.04, the best performer, by likely aligning SHORT entries with the actual bearish trend and using social/forum sentiment to filter high-conviction trades. gptoss_20b_simple: Moderate trade frequency (149 t... Confidence : 75% Total samples : 149 Times confirmed : 1 First seen : 2026-02-01 Details : gptoss_20b_simple: Moderate trade frequency (149 trades) with selective position management, achieving small positive PnL (+$8.60) by avoiding counter-trend trades and cutting losers early. gptoss_20b_simple achieves positive PnL (+$140.71)... Confidence : 75% Total samples : 48 Times confirmed : 1 First seen : 2026-01-30 Details : gptoss_20b_simple achieves positive PnL (+$140.71) with 48 trades using disciplined position closing: 'Close the small profitable short to lock in gains' - taking profits on winning shorts rather than holding. gpt_simple's extremely low trade frequency (10 tra... Confidence : 75% Total samples : 10 Times confirmed : 1 First seen : 2026-02-02 Details : gpt_simple's extremely low trade frequency (10 trades/24h) with selective entries limits losses to -$140.76 despite wrong-way positioning. Quick loss-cutting on BTC short ('BTC is grinding higher against my short') demonstrates adaptive behavior. Conservative trading (7 trades/24h) limits losses ... Confidence : 75% Total samples : 7 Times confirmed : 1 First seen : 2026-01-31 Details : Conservative trading (7 trades/24h) limits losses in uncertain markets. gpt_simple's minimal activity resulted in only -$8.44 loss vs much larger losses from aggressive traders. Zero trading preserves capital perfectly when mark... Confidence : 75% Total samples : 2 Times confirmed : 1 First seen : 2026-03-06 Details : Zero trading preserves capital perfectly when market direction is uncertain or agent lacks conviction. gptoss_skill_aware (0 trades, $0) and qwen35_skill_aware (0 trades, $0) avoided all losses by staying out entirely. Agentic multi-step reasoning with moderate trade f... Confidence : 72% Total samples : 30 Times confirmed : 1 First seen : 2026-03-08 Details : Agentic multi-step reasoning with moderate trade frequency (30 trades) achieved +$123.26 in a mixed market. gptoss_agentic's sophisticated analysis appears to have correctly navigated the mixed regime with selective entries. Low trade frequency with correct directional aware... Confidence : 72% Total samples : 27 Times confirmed : 1 First seen : 2026-03-09 Details : Low trade frequency with correct directional awareness. gptoss_journal_aware's 27 trades (+$55.48) vs gptoss_skill_aware's 31 trades (-$340.51) shows that self-reflective management outperforms sophisticated validation frameworks when both trade at similar frequencies. Journal-aware self-reflective position management ... Confidence : 67% Total samples : 48 Times confirmed : 1 First seen : 2026-03-06 Details : Journal-aware self-reflective position management with high trade frequency (48 trades) in a uniformly bearish market. gptoss_journal_aware achieved +$296.01, the best performer, suggesting meta-cognitive exits ('close to reduce exposure', 'we are overtrading') correctly captured short-side profits and limited drawdowns when the market moved -3% to -5%. Low trade frequency with self-reflective or skill-... Confidence : 65%
Keywords that activate this skill. Click one to copy it.

This skill does not provide trigger words.

The downloaded .skill package contains the following fields.
Field Description
formatFormat tag (skill/v1)
skill_idUnique skill ID
nameSkill name
versionVersion
descriptionDescription
categoryCategories (array)
trigger_wordsTrigger words
tagsTags
sourceSource
source_urlSource URL (this page)
exported_atExported at (set per download)
system_promptSystem prompt body
model_configModel config: provider / model / temperature / max_tokens / top_p
examplesExamples
install_guideImport guide for Coze / Dify / Claude / custom frameworks
The same skill can be exported in different platform formats.
.skill Standard format with system_prompt and model_config, ready for any agent framework Download
.skillpro Enhanced format with scripts, tools, dependencies and hooks Download
.json Plain JSON export with system_prompt and model parameters only Download
Coze Markdown with frontmatter, for Coze platform import Download
Dify Dify DSL, import directly after creating an app Download

每日精选 Skill 推荐,免费送到你邮箱

输入邮箱,每天接收一个精选 AI Agent 技能推荐。完全免费,持续更新。

提交后我们会发送一封确认邮件,点击邮件里的链接才会开始收信。

完全免费,取消任意时间。我们不会发送垃圾邮件。