Loading...
Loading...
Found 574 Skills
Quantitative signal scanning and position sizing tool based on the original Turtle Trading method. It retrieves market data for A-shares / Hong Kong stocks / US stocks / Singapore stocks via longbridge CLI, and automatically calculates ATR (N value), breakout signals (System 1 / System 2), stop-loss prices, add-on positions, and Unit position sizes. Trigger this tool when users mention 海龟, turtle, 海龟交易, 海龟信号, turtle signal, turtle trading, or ask about breakout signals, ATR, N value, Unit positions, stop-loss prices, add-on positions, S1/S2 signals, 20-day high/low, 55-day breakout, or request to scan watchlists/indexes for trading signals using the turtle system. It also triggers when users say "扫描突破信号", "帮我算Unit", "海龟止损", "海龟系统分析", or any combination of a stock name/code with "海龟". **Applicable scenarios:** - Scan for breakout signals (20-day/55-day high/low breakouts) after daily market close - Calculate ATR, stop-loss prices, and add-on positions for single stocks or batches of targets - Calculate reasonable Unit position sizes based on account net assets - Determine whether existing positions trigger exit or add-on conditions - Scan turtle signals for watchlist stocks / index components **Not applicable for:** - Fundamental analysis (Turtle system is purely technical) - Predicting price direction - Automatic order placement (only outputs signals; users operate on their own) - Short-selling opening operations for A-shares/Hong Kong stocks/Singapore stocks
Use this skill when a user provides a torrent name or file name and wants to fix recognition issues, or asks to add/manage custom identifiers (自定义识别词). This skill generates identifier rules based on the WordsMatcher preprocessing logic, checks for duplicates against existing rules, and saves them via MCP tools. Because custom identifiers are global, generated rules must default to conservative, sample-specific regex patterns instead of broad matches unless the user explicitly wants global cleanup. Applicable scenarios include: 1) A torrent or file name is incorrectly recognized (wrong title, season, episode, etc.); 2) The user wants to block unwanted keywords from torrent names; 3) The user needs episode offset rules for series with non-standard numbering; 4) The user wants to force recognition of a specific media by TMDB/Douban ID.
Build and operate predictive models for logistics networks—demand forecasting at SKU/location/lane granularity; inventory positioning and safety stock optimization interfaces; ETA and lead-time prediction; capacity and congestion signals; route and network flow forecasting at model-integration level; cold chain and perishables; promotion and seasonality; model monitoring, drift, and backtesting against operational KPIs (fill rate, OTIF, WMAPE/MAPE). Use for predictive logistics, demand forecasting logistics, ETA prediction, inventory positioning, safety stock optimization, OTIF forecast, lane demand, WMAPE, logistics ML, capacity forecasting logistics, or cold chain forecast—not pure OR/MIP without logistics domain (operations-research-algorithm-developer), supply chain strategy only (supply-chain-manager), WMS feature dev (wms-developer), fleet telematics ingestion (geospatial-telematics-developer), generic ML without logistics (data-scientist), or EDI document mapping (edi-engineer).
Run fable-mode execution discipline on Claude Opus — the strongest staged run available. Routes the task to the @fable-orchestrator agent (Opus, Write-less), which stages the work, delegates artifact production to @fable-worker-sonnet / @fable-worker-haiku, and cold-checks deliverables with @fable-verifier. Trigger when the user explicitly asks for thorough/systematic/"deep work" handling on the strongest model ("fable on opus", "stage this on opus", "deep work mode, opus"). Do NOT use for ordinary single-pass tasks — and prefer fable-sonnet or fable-haiku when the task doesn't need peak reasoning.
Launch an intelligent sub-agent with automatic model selection based on task complexity, specialized agent matching, Zero-shot CoT reasoning, and mandatory self-critique verification
Run /debug to find and fix a bug's root cause: a test failing for an unclear reason, /check verify finding a failure, or behavior being wrong. Runs a reproduce, localize, hypothesize, test, fix, verify loop, makes the minimal fix, and hands a regression test to /test. No features, no extra refactors.
Facilitates the fifth step of a proven ideal-customer (ICP) method: mapping inciting events — the specific trigger moments that move a perfect-fit customer from could-buy-someday to buying-today. Takes a keystones file (K1, K2, … with market segments) and, when available, customer-interview findings; walks the keystones one at a time, harvesting real trigger stories from interview evidence (marked observed) and working backward through brainstorm lenses — crises, seasonal cycles, strategic windows, personal life-changes — for the rest (marked hypothesized), recording each event with the keystone it couples to and how to find prospects in that condition, in INCITING-EVENTS.md (E1, E2, …). Load when the user has keystones and asks what makes customers buy now, what triggers a purchase, or 'run the inciting-events step.' Do NOT load to derive keystones or deal-breakers (previous steps), to write the final ideal-customer definition (next step), or to write the ads themselves.
Apply causal inference whenever the user is interpreting metrics, debugging system behavior, reading A/B test results, or trying to understand whether an observed change was caused by an action or by something else. Triggers on phrases like "X caused Y", "since we deployed this, metrics changed", "the A/B test showed a lift", "why did this metric move?", "is this correlation or causation?", "we changed X and Y improved", "how do we know this worked?", "the data shows…", or any situation where conclusions are being drawn from observational data. Also trigger before any decision based on metric interpretation — confusing correlation with causation leads to interventions that don't work and misattribution of credit. Never assume causation without applying this skill.
Reflective sleep-and-dream heuristic for learning from recent experience. Use when the user asks to sleep on something, dream about it, reflect overnight, learn from yesterday, or extract lessons after a meaningful task, conversation, or debugging session. Avoid for first-pass analysis, simple factual lookups, direct execution, or tasks that do not benefit from reflection.
Framework for building LLM-powered applications with agents, chains, and RAG. Supports multiple providers (OpenAI, Anthropic, Google), 500+ integrations, ReAct agents, tool calling, memory management, and vector store retrieval. Use for building chatbots, question-answering systems, autonomous agents, or RAG applications. Best for rapid prototyping and production deployments.
Builds a week-by-week Q4 restock plan from August through November with FBA inbound delay buffers, peak velocity multipliers, and FBM fallback triggers. Q4 is 30-40% of annual revenue and FBA receiving takes 2-3 weeks during peak. Standard restock math stocks out at the worst possible moment. Use when a user asks about Q4 planning, Black Friday inventory, or Prime Big Deal Days restock. Trigger phrases: "Q4 restock plan", "Black Friday inventory", "Prime Big Deal Days restock", "Christmas Amazon inventory", "FBM fallback for FBA". Works with zero tools.
Graph of Thoughts (GoT) Controller - 管理研究图状态，执行图操作（Generate, Aggregate, Refine, Score），优化研究路径质量。当研究主题复杂或多方面、需要策略性探索（深度 vs 广度）、高质量研究时使用此技能。