Xu Xiaoming Stock Trading Strategy Agent Skill
Applies Xu Xiaoming's A-share technical analysis methodology to assess market regimes, evaluate trend-structure signals, and generate…
- Context cost
- 993.4k tokensestimated from the bundle; more than a 200k window holds
- Bundle
- 38 of 187 fileshanded out here; the rest is at its source
- Licence
- not declarednone named by its author
- Last change
- —no release on file
- Servers it uses
- Noneruns standalone
What it does
徐小明技术分析思想 —— 蒸馏自公众号「投资明见」2019-2026年2124篇文章。 四核心模型（两策略轮换+循环哲学+工具简化+"不做"体系），三时期分化（完备→收缩→极简）， 六市况阶段内外对照。四层架构：能力层（怎么想）+ 表达层（怎么说）+ 规则层（什么规则）+ 工具接入层（怎么实现）。 规则层框架无关，用户可替换工具。默认提供三合一+CHOP+MACD结构+九转序列+五指数共振+分钟线结构+NH/NL广度+博客对照完整实现。 触发词：「徐小明」「小明技术分析」「趋势为王结构修边」「用徐小明视角」「看一下盘面」「盘中分析」。 用于：A股技术分析操作哲学、市况判断与策略匹配、收盘全流程报告（九节）、盘中实时简报（七段）、博客独立对照验证。 不用于：选股、基本面分析、宏观择时、日内高频交易（分钟线/分时图操作）、板块轮动分析。盘中分析判断尺度为日线级别，非日内高频。
Installed, it changes the agent in these ways.
What this skill changes about the agent has not been set out here as a list. Its SKILL.md, shown below, says it in the author’s own words.
And these are the jobs people reach for it for.
The skill itself
A skill is instructions the model reads, so you can see what you are getting: below is the start of its SKILL.md. The front matter loads with every session, the body loads when the skill triggers, and the whole file comes with the install.
Nothing in a skill executes. The client loads the text and the model follows it, so a skill can be audited the way a runbook is — by reading it.
What it costs in context
Skills are not billed by the call. They are paid for in context: every token the instructions occupy is a token your code, your diff and your conversation cannot use. Here is what this one takes and when it takes it.
993.4k tokens, estimated from the bundle at four bytes to the token — more than the whole window, so it cannot all be held at once: the model reads SKILL.md when the skill fires and the other files only as the instructions call for them. Heavy. Teams tend to install this one per project rather than globally, and load it only when the job comes up.
A server charges by the month. A skill charges once per session, in context, and then keeps charging it for as long as the session lives.
Before and after
The same question, put to the same model twice: once as it comes, and once with these instructions loaded.
No worked example has been published for this skill yet.
The procedure it runs
The procedure has not been set out here step by step. It is in the body of its SKILL.md: the start of it is shown above, and the whole file comes with the install.
These steps are written for a model to follow, not executed by a runtime. It can still be told to skip one, and it will say so when it does.
Servers it uses
None. This skill calls no MCP servers at all.
Everything it needs is in the instructions, so it works in a project with nothing connected — the model reads the file and changes how it works with what it can already reach.
Its SKILL.md does not declare which tools it asks the agent for, so what it touches is whatever the instructions themselves ask for — read them before you install it.
This SKILL.md carries no allowed-tools line, so neither is declared. A skill grants no permissions of its own — it can only ask for tools your client already has.
What it will not do
Every skill is narrow, and the useful ones say where they stop. These are the jobs this one is the wrong tool for.
What this skill is not for has not been published here. Nothing is implied by that: it is a section the author has not filled in.
What is in the bundle
The download carries 38 of the 187 files in its source folder; the list here names 38 files, 4091.7 kB on disk. Mostly text — the instructions the model reads — with 6 scripts in it that your client would run only if the instructions tell it to. The other 149 are not in the download, nor are at least 1 file of types a skill download does not carry — the first 100 are listed after these, with why.
- SKILL.md142.7 kB
- _backtest_monthly.py5.9 kB
- execution-checklist.md5.8 kB
- inference_signals.py23.0 kB
- README.md7.1 kB
- realtime.py34.1 kB
- strategy-output-template.md21.3 kB
- verdict_engine.py14.0 kB
- verdict_v7.py32.1 kB
- verdict_v441.py9.0 kB
- data/breadth_daily.csv47.2 kB
- data/daily_raw.csv943.2 kB
- data/daily_spot_cache.json4.1 kB
- data/minute_raw_30_sh000001_上证指数.csv132.3 kB
- data/minute_raw_30_sh000300_沪深300.csv132.1 kB
- data/minute_raw_30_sh000688_科创50.csv129.3 kB
- data/minute_raw_30_sh000905_中证500.csv132.0 kB
- data/minute_raw_30_sz399001_深证成指.csv140.0 kB
- data/minute_raw_30_sz399006_创业板指.csv132.0 kB
- data/minute_raw_60_sh000001_上证指数.csv133.2 kB
- data/minute_raw_60_sh000300_沪深300.csv132.2 kB
- data/minute_raw_60_sh000688_科创50.csv127.6 kB
- data/minute_raw_60_sh000905_中证500.csv132.2 kB
- data/minute_raw_60_sz399001_深证成指.csv138.8 kB
- data/minute_raw_60_sz399006_创业板指.csv132.3 kB
- data/minute_raw_90_sh000001_上证指数.csv76.2 kB
- data/minute_raw_90_sh000300_沪深300.csv75.9 kB
- data/minute_raw_90_sh000688_科创50.csv73.2 kB
- data/minute_raw_90_sh000905_中证500.csv75.9 kB
- data/minute_raw_90_sz399001_深证成指.csv79.9 kB
- data/minute_raw_90_sz399006_创业板指.csv75.9 kB
- data/minute_raw_120_sh000001_上证指数.csv105.2 kB
- data/minute_raw_120_sh000300_沪深300.csv104.2 kB
- data/minute_raw_120_sh000688_科创50.csv100.1 kB
- data/minute_raw_120_sh000905_中证500.csv104.2 kB
- data/minute_raw_120_sz399001_深证成指.csv109.1 kB
- data/minute_raw_120_sz399006_创业板指.csv104.2 kB
- data/minute_structure_v2_cyb.csv124.2 kB
Not in the download
The first 100 are named here, those the skill reads first. Not carried is a type a skill download does not hold and over the limit is past 200 files or 4 MB a skill. Marked read: its SKILL.md or scripts name the file, so without it the skill works only in part. Take them from its source.
- data/minute_structure_v2_sh.csvread · over the limit
- data/minute_structure_v2_sz.csvread · over the limit
- data/structure_signals.csvread · over the limit
- data/turn_sequence_events.csvread · over the limit
- data/verdict_v7.csvread · over the limit
- data_layer/breadth.pyread · over the limit
- data_layer/build.pyread · over the limit
- data_layer/fetch.pyread · over the limit
- data_layer/loader.pyread · over the limit
- data_layer/process.pyread · over the limit
- data_layer/realtime.pyread · over the limit
- references/2635-reality-check.mdread · over the limit
- references/blind-test-methodology.mdread · over the limit
- references/blind-test-v5-results.mdread · over the limit
- references/blind-test-v6-2023-2024.mdread · over the limit
- references/blind-test-v6-annual.mdread · over the limit
- references/blind-test-v7-2020-2022.mdread · over the limit
- references/bottom-structure-dual-pattern.mdread · over the limit
- references/channel-classifier-methodology.mdread · over the limit
- references/composite-verdict-framework.mdread · over the limit
- references/cyb-kc-resonance-backtest.mdread · over the limit
- references/divergence-backtest-v4611.mdread · over the limit
- references/divergence-evolution.mdread · over the limit
- references/downward-origin-oscillation-model.mdread · over the limit
- references/execution-checklist.mdread · over the limit
- references/h11-h12-during-unconfirmed-exit.mdread · over the limit
- references/hysteresis-backtest.mdread · over the limit
- references/inference-signals-backtest.mdread · over the limit
- references/llm-extraction-method.mdread · over the limit
- references/minute_methodology_final.mdread · over the limit
- references/minute-structure-index-map-pitfall.mdread · over the limit
- references/model-vs-xuxiaoming-divergence.mdread · over the limit
- references/operations-layer.mdread · over the limit
- references/original-text-dual-pattern.mdread · over the limit
- references/osc-standards-backtest.mdread · over the limit
- references/oscillation-debunked.mdread · over the limit
- references/oscillation-origin-framework.mdread · over the limit
- references/oscillation-to-bearish-transition.mdread · over the limit
- references/pitfalls.mdread · over the limit
- references/post-close-data-pipeline.mdread · over the limit
- references/pure-structure-chop-trajectory.mdread · over the limit
- references/pure-structure-entry-characteristics.mdread · over the limit
- references/r8-double-low-confirm-backtest.mdread · over the limit
- references/r8-prep-during-wait-backtest.mdread · over the limit
- references/realtime-data-sources.mdread · over the limit
- references/resonance-stratification.mdread · over the limit
- references/rule-appendix-original-texts.mdread · over the limit
- references/rule-engine-hierarchy.mdread · over the limit
- references/s1-analysis-上行入震.mdread · over the limit
- references/signal-layer-priority.mdread · over the limit
- references/strategy-output-template.mdread · over the limit
- references/structure-is-a-process-924-case.mdread · over the limit
- references/time-symmetry-appendix.mdread · over the limit
- references/time-symmetry-backtest.mdread · over the limit
- references/trend-quantification-tools.mdread · over the limit
- references/upward-origin-bs-gap.mdread · over the limit
- references/upward-origin-oscillation-model.mdread · over the limit
- references/v7-architecture.mdread · over the limit
- references/verdict-v6-chop-framework.mdread · over the limit
- references/verdict-v4611-architecture.mdread · over the limit
- references/wave-theory-appendix.mdread · over the limit
- reports/2026-07-10-patch-minute-structure.mdread · over the limit
- scripts/backtest_v5.pyread · over the limit
- scripts/blind_test_compare.pyread · over the limit
- scripts/blind_test_sample.pyread · over the limit
- scripts/blog_monitor.pyread · over the limit
- scripts/channel_classifier_xu.pyread · over the limit
- scripts/check_a20.pyread · over the limit
- scripts/chop_threshold_backtest.pyread · over the limit
- scripts/coordination.pyread · over the limit
- scripts/failure_tracker.pyread · over the limit
- scripts/full_validation.pyread · over the limit
- scripts/inference_signals.pyread · over the limit
- scripts/merge_daily_state.pyread · over the limit
- scripts/minute_structure_v2.pyread · over the limit
- scripts/multi_index_resonance.pyread · over the limit
- scripts/sequence_aftermath.pyread · over the limit
- scripts/sequence_engine.pyread · over the limit
- scripts/sequence_failure_analysis.pyread · over the limit
- scripts/strategy-output-template.mdread · over the limit
- scripts/structure_engine.pyread · over the limit
- scripts/trajectory.pyread · over the limit
- scripts/unimportant_detector.pyread · over the limit
- scripts/verdict_engine.pyread · over the limit
- scripts/verdict_v7.pyread · over the limit
- scripts/verdict_v441.pyread · over the limit
- .gitignorenot carried
- data/minute_structure_v2_kc.csvover the limit
- data/seal_test_results.jsonover the limit
- references/blind-test-report.mdover the limit
- references/blind-test-v7-2023-2024.mdover the limit
- references/blind-test-v7-results.mdover the limit
- references/breadth-backtest-2025-spring.mdover the limit
- references/case-20260713-model-vs-blog.mdover the limit
- references/chop-sustained-threshold.mdover the limit
- references/context-window-fix.mdover the limit
- references/cross-analysis-findings.mdover the limit
- references/data-contract-turn-sequence-events.mdover the limit
- references/divergence-xxm-backtest.mdover the limit
- references/down-origin-shock-entry-path.mdover the limit
A skill installs nothing and depends on nothing: it is a folder your client reads. This one carries 6 scripts beside the text, so the download is 38 files — not the whole of its source: at least 150 files are left out, and the skill reads 86 of them. The first 100 are listed under the files, with why.
Install
Installing copies the bundle into your project. Nothing runs at install time — the files sit on disk until the model reads them.
# Xu Xiaoming Stock Trading Strategy · 993.4k tokens when loaded · macOS, Linux
mkdir -p .claude/skills && t=$(mktemp -d) &&
curl -fsSL -o "$t/bundle.tar.gz" https://mcprush.com/api/skills/xu-xiaoming-stock-trading-strategy/bundle.tar.gz &&
tar -xzf "$t/bundle.tar.gz" -C "$t" && [ -d "$t/xu-xiaoming-stock-trading-strategy" ] &&
{ [ ! -e .claude/skills/xu-xiaoming-stock-trading-strategy ] || { mv .claude/skills/xu-xiaoming-stock-trading-strategy "$t/.previous" && echo "Previous copy moved to $t/.previous"; }; } &&
mv "$t/xu-xiaoming-stock-trading-strategy" .claude/skills/ && echo "Installed .claude/skills/xu-xiaoming-stock-trading-strategy"npx mcprush@latest skill add tadalafil-cure/xu-xiaoming-stock-trading-strategyWindows PowerShell says 'running scripts is disabled on this system'? Run Set-ExecutionPolicy -Scope CurrentUser RemoteSigned once, or type npx.cmd instead of npx (codex.cmd, gemini.cmd, grok.cmd likewise). PowerShell 7 needs neither. Writes to .claude/skills/xu-xiaoming-stock-trading-strategy/ in the current project. For every project instead, run mkdir -p ~/.claude/skills && t=$(mktemp -d) && curl -fsSL -o "$t/bundle.tar.gz" https://mcprush.com/api/skills/xu-xiaoming-stock-trading-strategy/bundle.tar.gz && tar -xzf "$t/bundle.tar.gz" -C "$t" && [ -d "$t/xu-xiaoming-stock-trading-strategy" ] && { [ ! -e ~/.claude/skills/xu-xiaoming-stock-trading-strategy ] || { mv ~/.claude/skills/xu-xiaoming-stock-trading-strategy "$t/.previous" && echo "Previous copy moved to $t/.previous"; }; } && mv "$t/xu-xiaoming-stock-trading-strategy" ~/.claude/skills/ && echo "Installed ~/.claude/skills/xu-xiaoming-stock-trading-strategy" (macOS, Linux) — or add --global to the npx line. The curl line replaces .claude/skills/xu-xiaoming-stock-trading-strategy/ only after the download succeeds, so files left there by another skill of the same name, or by an older release, do not mix in. A copy already there is moved aside, not deleted, and the line prints where. It appears as XuXiaoming-StockTradingStrategy in Gemini CLI, Codex, Copilot and Grok (its own name); Claude Code lists it by the folder name, xu-xiaoming-stock-trading-strategy. Not in this download: at least 150 files of its source, 86 of them read by its SKILL.md or scripts (the list, and why). Take them from its source.
Which clients pick it up on their own
A skill is a folder of text. Most of these read a skills folder and pick it up without being told; the rest take the same folder another way — an upload in the app, a parameter on an API call — and then choose it by its description just the same.
Versions
Its author publishes no version number, so there is nothing here to pin to: what you install is the folder as it stands today. Instructions change more often than APIs do — a skill can be rewritten entirely without anything it depends on moving.
Nothing to pin to: this skill carries no version number of its own. What you install is what the folder holds on the day you install it.
Reviews
Nobody has reviewed this skill yet. The rating is the mean of the reviews written here, so there is none until somebody writes the first.
A review is tied to an account so it can be answered, edited and taken down by the person who wrote it.
Sign inAnybody with an account, one review each. It carries the version that was current when you wrote it, and the publisher may reply once.