Matplotlib
zLanqing/codex-claude-academic-skills
Low-level plotting library for full customization. An agent skill from zLanqing/codex-claude-academic-skills.
QuantMind 平台运营操作技能 — 覆盖模型训练、模型管理、后台数据更新、字段信息查询、RSS 新闻对接与分析。在 QuantBot / Claude Code 中处理模型训练、数据同步、新闻分析等任务时使用。触发词:模型管理、数据更新、字段信息、RSS、新闻分析、查看数据、同步数据
$ npx skills add qusong0627/QuantMind --skill quantmind-operations -a claude-codeProject install by default; add -g for ~/.claude/skills/.
$ gh skill install qusong0627/QuantMind quantmind-operations --agent claude-codeProject scope by default; add --scope user for a personal install. Needs GitHub CLI 2.90.0 or later (public preview).
$ git clone --depth 1 https://github.com/qusong0627/QuantMind.git skills-src && mkdir -p .claude/skills && cp -r skills-src/skills/quantmind-operations .claude/skills/quantmind-operations && rm -rf skills-srcUse ~/.claude/skills/ instead of .claude/skills for a personal install. The folder must contain SKILL.md.
Claude Code skills documentation · loads skills from .claude/skills/
Install the "quantmind-operations" agent skill from https://github.com/qusong0627/QuantMind/tree/master/skills/quantmind-operations into .claude/skills/quantmind-operations/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "quantmind-operations", then confirm the skill loads.Claude Code copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$skill-installer install https://github.com/qusong0627/QuantMind/tree/master/skills/quantmind-operationsType this inside Codex. $skill-installer <name> installs a curated skill from openai/skills. The installer writes to $CODEX_HOME/skills (default ~/.codex/skills). Restart Codex if the skill does not show up.
$ npx skills add qusong0627/QuantMind --skill quantmind-operations -a codexProject install goes to .agents/skills/; add -g for ~/.codex/skills/.
$ gh skill install qusong0627/QuantMind quantmind-operations --agent codexProject scope by default (.agents/skills/); add --scope user for a personal install.
$ git clone --depth 1 https://github.com/qusong0627/QuantMind.git skills-src && mkdir -p .agents/skills && cp -r skills-src/skills/quantmind-operations .agents/skills/quantmind-operations && rm -rf skills-srcUse ~/.agents/skills/ instead of .agents/skills for a personal install.
Codex skills documentation · loads skills from .agents/skills/
Install the "quantmind-operations" agent skill from https://github.com/qusong0627/QuantMind/tree/master/skills/quantmind-operations into .agents/skills/quantmind-operations/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "quantmind-operations", then confirm the skill loads.Codex copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$ npx skills add qusong0627/QuantMind --skill quantmind-operations -a cursorProject install goes to .agents/skills/; add -g for ~/.cursor/skills/.
$ gh skill install qusong0627/QuantMind quantmind-operations --agent cursorProject scope by default (.agents/skills/); add --scope user for a personal install.
$ git clone --depth 1 https://github.com/qusong0627/QuantMind.git skills-src && mkdir -p .cursor/skills && cp -r skills-src/skills/quantmind-operations .cursor/skills/quantmind-operations && rm -rf skills-srcUse ~/.cursor/skills/ instead of .cursor/skills for a personal install.
Cursor skills documentation · loads skills from .cursor/skills/, .agents/skills/, .claude/skills/, .codex/skills/
Install the "quantmind-operations" agent skill from https://github.com/qusong0627/QuantMind/tree/master/skills/quantmind-operations into .cursor/skills/quantmind-operations/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "quantmind-operations", then confirm the skill loads.Cursor copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$ gemini skills install https://github.com/qusong0627/QuantMind.git --path skills/quantmind-operations--scope user (default) or --scope workspace; --path is the subfolder of the repo that holds the skill; --consent skips the security confirmation prompt.
$ npx skills add qusong0627/QuantMind --skill quantmind-operations -a gemini-cliProject install goes to .agents/skills/; add -g for ~/.gemini/skills/.
$ gh skill install qusong0627/QuantMind quantmind-operations --agent gemini-cliProject scope by default (.agents/skills/); add --scope user for a personal install.
$ git clone --depth 1 https://github.com/qusong0627/QuantMind.git skills-src && mkdir -p .gemini/skills && cp -r skills-src/skills/quantmind-operations .gemini/skills/quantmind-operations && rm -rf skills-srcUse ~/.gemini/skills/ instead of .gemini/skills for a personal install, then run /skills reload.
Gemini CLI skills documentation · loads skills from .gemini/skills/, .agents/skills/
Install the "quantmind-operations" agent skill from https://github.com/qusong0627/QuantMind/tree/master/skills/quantmind-operations into .gemini/skills/quantmind-operations/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "quantmind-operations", then confirm the skill loads.Gemini CLI copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$ gh skill install qusong0627/QuantMind quantmind-operationsInstalls for Copilot at project scope by default; add --scope user for a personal install. Preview a skill first with gh skill preview. Needs GitHub CLI 2.90.0 or later (public preview).
$ npx skills add qusong0627/QuantMind --skill quantmind-operations -a github-copilotProject install goes to .agents/skills/; add -g for ~/.copilot/skills/.
$ git clone --depth 1 https://github.com/qusong0627/QuantMind.git skills-src && mkdir -p .github/skills && cp -r skills-src/skills/quantmind-operations .github/skills/quantmind-operations && rm -rf skills-srcUse ~/.copilot/skills/ instead of .github/skills for a personal install. Commit .github/skills so cloud agent and code review can use it.
GitHub Copilot skills documentation · loads skills from .github/skills/, .claude/skills/, .agents/skills/
Install the "quantmind-operations" agent skill from https://github.com/qusong0627/QuantMind/tree/master/skills/quantmind-operations into .github/skills/quantmind-operations/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "quantmind-operations", then confirm the skill loads.GitHub Copilot copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$ npx skills add qusong0627/QuantMind --skill quantmind-operations -a opencodeOpenCode documents no install command of its own. Project install goes to .agents/skills/; add -g for ~/.config/opencode/skills/.
$ gh skill install qusong0627/QuantMind quantmind-operations --agent opencodeProject scope by default (.agents/skills/); add --scope user for a personal install.
$ git clone --depth 1 https://github.com/qusong0627/QuantMind.git skills-src && mkdir -p .opencode/skills && cp -r skills-src/skills/quantmind-operations .opencode/skills/quantmind-operations && rm -rf skills-srcUse ~/.config/opencode/skills/ instead of .opencode/skills for a personal install.
OpenCode skills documentation · loads skills from .opencode/skills/, .claude/skills/, .agents/skills/
Install the "quantmind-operations" agent skill from https://github.com/qusong0627/QuantMind/tree/master/skills/quantmind-operations into .opencode/skills/quantmind-operations/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "quantmind-operations", then confirm the skill loads.OpenCode copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
quantmind-operationsQuantMind 平台运营操作技能 — 覆盖模型训练、模型管理、后台数据更新、字段信息查询、RSS 新闻对接与分析。在 QuantBot / Claude Code 中处理模型训练、数据同步、新闻分析等任务时使用。触发词:模型管理、数据更新、字段信息、RSS、新闻分析、查看数据、同步数据
Quantmind Operations is an agent skill from qusong0627/QuantMind. QuantMind 平台运营操作技能 — 覆盖模型训练、模型管理、后台数据更新、字段信息查询、RSS 新闻对接与分析。在 QuantBot / Claude Code 中处理模型训练、数据同步、新闻分析等任务时使用。触发词:模型管理、数据更新、字段信息、RSS、新闻分析、查看数据、同步数据
Its SKILL.md is about 4k tokens, which your agent loads only when the skill is triggered. The skill folder holds 3 other files, including reference files (for example `references/news-analysis.md` and `references/training-data-ops.md`).
It sits in Data & Analytics. The repository describes itself as: QuantMind(量化大脑)开源版是一款面向个人开发者与投研团队的 AI 原生多市场量化交易平台。深度集成微软 Qlib、RD-Agent 因子演化与 QuantBot全能工作台,提供从 300+ 维因子挖掘、13 种机器学习与深度学习模型工场、Qlib 高性能回测、截面批量推理、7x24… The licence is AGPL-3.0.
9 steps, taken from the step headings in SKILL.md.
Read from SKILL.md and the folder at commit 2e93d9a. It shows what the files ask for, not the result of running them.
Pre-approves nothing: there is no allowed-tools line, so your agent's usual permission prompts apply.
From allowed-tools in the SKILL.md frontmatter.
Shell commands in SKILL.md call:
curlpythonpython3From the folder's file list and the shell code blocks in SKILL.md.
No URLs in SKILL.md. Its commands use curl, which can reach the network depending on how they are called.
From URLs in SKILL.md, links to its own repository left out.
Names no API keys, tokens, secrets or passwords.
From names ending in _API_KEY, _TOKEN, _SECRET, _KEY or _PASSWORD in SKILL.md.
Quantmind Operations loads about 4k tokens when it runs, and up to ~6k if it reads all its reference files. Until then it costs about 42 tokens; SKILL.md has 545 words of instructions outside code blocks.
Estimates: characters ÷ 4, the usual rule of thumb; real counts depend on the model's tokenizer. Scripts and assets cost tokens only if the agent reads them.
The automated check found no risky patterns in SKILL.md.
Automated static check — not a guarantee. Review scripts before installing. It scans the text of SKILL.md for risky patterns (piping downloads into a shell, reading credential files, hidden Unicode, destructive commands); files beside SKILL.md are not scanned.
The full file from qusong0627/QuantMind at commit 2e93d9a, republished under its AGPL-3.0 licence (© qusong0627). 545 words, ~3,994 tokens.
.claude/skills/quantmind-operations/SKILL.md (or your agent's skills folder). This skill also uses 2 other files; get the full folder from GitHub.⚙️ 本技能遵循公共运行环境契约(最高优先级,先于本文其余内容执行): 详见 _shared/env-contract.md,执行前先读它。
QuantMind 量化平台的完整运营操作指南。所有 API 都通过 API 网关(默认 http://127.0.0.1:8000 或 http://192.168.31.68:3080)访问,统一加 /api/v1 前缀。
所有请求需要 Bearer Token:
# 获取 token(admin 账号)
TOKEN=$(curl -s -X POST $BASE/api/v1/auth/login \
-H "Content-Type: application/json" \
-d '{"username":"admin","password":"admin123","tenant_id":"default"}' \
| python3 -c "import sys,json; print(json.load(sys.stdin).get('access_token',''))")
# 通用请求头
AUTH="Authorization: Bearer $TOKEN"
CT="Content-Type: application/json"模型训练分 5 步,与前端 ModelTrainingPage 一致:
特征选择 → 训练目标 → 参数配置 → 执行训练 → 结果入库# 获取特征字典(类别/数量由 QuantDB l1_factors 动态生成,以接口返回为准)
curl -s -H "$AUTH" "$BASE/api/v1/models/feature-catalog"
# 带数据覆盖统计(含建议训练/验证/测试区间)
curl -s -H "$AUTH" "$BASE/api/v1/models/feature-catalog?include_coverage=true"
# 管理端特征字典(含扫描详情)
curl -s -H "$AUTH" "$BASE/api/v1/admin/models/feature-catalog"选择特征 key 列表(如 ["mom_ret_5d", "vol_std_20"])或按类别(feature_categories)。
特征类别由后端特征目录动态生成,随 QuantDB l1_factors 数据版本变化(示例 version 20260831 返回 10 类 110 特征:momentum / fundamental / money_flow / style / technical / turnover / concept / volatility / chip / industry)。先读接口返回的 categories[].id,不要硬编码类别清单。
target_horizon_days:预测周期(T+1 / T+5 / T+20 等)target_mode:regression(回归)或 classification(分类)label_formula:标签计算公式(如 close_future/close - 1)effective_trade_date:生效交易日training_window:训练窗口(如 rolling)train_start/end、valid_start/end、test_start/end、val_rationum_boost_round、early_stopping_rounds、lgb_params/xgb_params/catboost_params/dl_paramscontext:initial_capital、benchmark、commission_rate、slippage、deal_price、market、industry_as_featurecurl -s -X POST "$BASE/api/v1/models/run-training" -H "$AUTH" -H "$CT" -d '{
"model_type": "lightgbm",
"model_types": ["lightgbm", "xgboost", "catboost"],
"ensemble": "stacking",
"job_name": "我的模型",
"display_name": "我的模型",
"train_start": "2022-01-01",
"train_end": "2024-12-31",
"valid_start": "2023-06-01",
"valid_end": "2024-06-30",
"test_start": "2024-07-01",
"test_end": "2024-12-31",
"val_ratio": 0.15,
"num_boost_round": 1000,
"early_stopping_rounds": 100,
"features": ["mom_ret_5d", "vol_std_20"],
"feature_categories": ["momentum", "volatility"],
"target_horizon_days": 1,
"target_mode": "regression",
"label_formula": "close_future/close - 1",
"effective_trade_date": "2025-01-02",
"training_window": "rolling",
"context": {"initial_capital": 1000000, "benchmark": "000300.SH", "commission_rate": 0.0003, "slippage": 0.001, "deal_price": "close", "market": "CN", "industry_as_feature": false},
"deploy_to_production": false
}'支持的 model_type(13 种,以 backend/shared/training/request.py::ALLOWED_MODEL_TYPES 为准):
lightgbm / xgboost / catboost / linear / random_forestgru / lstm / alstm / transformer / tabnet / tcn / nativetftmlp(sklearn 实现)hybrid_gru_tree 已剔除(QLIB map 无实现),勿再传(会被 422 拒绝/落入不支持)。ensemble 取值:none / stacking / blending / voting(多模型训练时生效)
可选高级参数:wfa(walk-forward,rolling/expanding)、target_horizon_days(单周期 T+N,1–30)、各模型专属超参 lgb_params/xgb_params/catboost_params/dl_params
⚠️ 已下线/死配置:
horizons(多周期,2026-09 随多周期训练一并清理)、optuna、n_folds(只建类型不参与序列化,传了不生效)。 返回:runId+ 有效/缺失特征统计
# 轮询训练状态(pending→running→completed/failed)
curl -s -H "$AUTH" "$BASE/api/v1/models/training-runs/{run_id}"
# 训练完成后模型进入 /models,可设为默认
curl -s -X PATCH -H "$AUTH" -H "$CT" "$BASE/api/v1/models/default" -d '{"model_id":"xxx"}'
# 查看模型列表确认入库
curl -s -H "$AUTH" "$BASE/api/v1/models"
curl -s -H "$AUTH" "$BASE/api/v1/models?include_archived=true"
# 系统内置模型
curl -s -H "$AUTH" "$BASE/api/v1/models/system-models"
# ⚠️ 手工融合模型已下线:原 ensemble 创建路由不存在
# (多周期 + 手工融合已于 2026-09 清理,见 backend/scripts/cleanup_multi_horizon_ensemble.py;
# 仅在多模型训练时用 model_types + ensemble 合成,或保留历史融合模型做推理兼容)curl -s -H "$AUTH" "$BASE/api/v1/admin/models/scan"# Qlib + 特征快照数据状态
curl -s -H "$AUTH" "$BASE/api/v1/admin/models/data-status"curl -s -H "$AUTH" "$BASE/api/v1/admin/models/precheck-inference"curl -s -X POST "$BASE/api/v1/admin/models/backtest" -H "$AUTH" -H "$CT" -d '{
"model_id": "xxx",
"start": "2024-01-01",
"end": "2024-12-31"
}'
# 可用回测日期
curl -s -H "$AUTH" "$BASE/api/v1/admin/models/backtest/trading-dates"
# 回测历史
curl -s -H "$AUTH" "$BASE/api/v1/admin/models/backtest/history/{model_id}"curl -s -X POST "$BASE/api/v1/admin/models/inference-backtest" -H "$AUTH" -H "$CT" -d '{
"model_id": "xxx"
}'# 提交同步任务,返回 task_id(market: A/CN=QuantDB, US=QuantUS, HK=QuantHK, BC=区块链, FUTURES=期货)
curl -s -X POST "$BASE/api/v1/admin/data-platform/daily-sync" -H "$AUTH" -H "$CT" -d '{
"market": "A",
"symbols": [],
"incremental": true,
"calibrate": true
}'
# 查询同步状态
curl -s -H "$AUTH" "$BASE/api/v1/admin/data-platform/daily-sync/status/{task_id}"各市场同步数据源:
| 市场 | market 值 | 数据源 | 说明 |
|---|---|---|---|
| A股 | A / CN | QuantDB SDK | 4阶段:parquet→PG→Qlib→特征快照 |
| 美股 | US | Yahoo Finance | quantus_daily_sync.py |
| 港股 | HK | Yahoo + akshare + CCASS | quanthk_daily_sync.py |
| 区块链 | BC | Binance | quantbc_daily_sync.py(支持 --minute) |
| 期货 | FUTURES | akshare | quantfutures_daily_sync.py |
# 查看全部市场定时配置
curl -s -H "$AUTH" "$BASE/api/v1/admin/data-platform/sync-schedule"
# 查看单市场配置
curl -s -H "$AUTH" "$BASE/api/v1/admin/data-platform/sync-schedule/{market}"
# 保存单市场配置(enabled/time/days/datasets/with_qlib)
curl -s -X POST -H "$AUTH" -H "$CT" "$BASE/api/v1/admin/data-platform/sync-schedule/{market}" \
-d '{"enabled": true, "time": "22:30", "days": [1,2,3,4,5], "datasets": ["all"], "with_qlib": true}'
# 立即触发一次同步(测试)
curl -s -X POST -H "$AUTH" "$BASE/api/v1/admin/data-platform/sync-schedule/{market}/run"curl -s -H "$AUTH" "$BASE/api/v1/admin/data-platform/sync-status"
curl -s -H "$AUTH" "$BASE/api/v1/admin/data-platform/sync-progress"# 同步数据集时带 with_qlib 触发 Qlib 缓存重建
curl -s -X POST -H "$AUTH" -H "$CT" "$BASE/api/v1/admin/data-platform/quantdb/sync-datasets" \
-d '{"datasets":["l1_factors","l2_factors"],"with_qlib":true}'
# 查看 Qlib + 特征快照数据状态(含年度快照详情)
curl -s -H "$AUTH" "$BASE/api/v1/admin/models/data-status"Qlib 路径(QlibDataBuilder.for_market):A股 .qlib_cache/cn_data,HK/US/BC/FUTURES 各目录下 .qlib_cache/{hk,us,bc,futures}_data。
# 指定年份(A股按年生成 model_features_{year}.parquet)
curl -s -X POST -H "$AUTH" "$BASE/api/v1/admin/data-platform/update-feature-parquet?year=2026"
# 多市场特征更新(非 A 股市场,market 必填:hong_kong / us_stock / crypto)
curl -s -X POST -H "$AUTH" "$BASE/api/v1/admin/data/update-market-features?market=hong_kong"
# 特征快照年度详情(A股逐年 metadata.json)
curl -s -H "$AUTH" "$BASE/api/v1/admin/models/data-status"特征快照结构:A股 db/feature_snapshots/model_features_{year}.parquet(含 .metadata.json 年度详情),非A股单体 model_features_{market}.parquet。
curl -s -X POST -H "$AUTH" "$BASE/api/v1/admin/data-platform/sync-fundamentals"
curl -s -H "$AUTH" "$BASE/api/v1/admin/data-platform/freshness"curl -s -H "$AUTH" "$BASE/api/v1/admin/data-platform/online-status"
curl -s -H "$AUTH" "$BASE/api/v1/admin/data-platform/sources"
curl -s -H "$AUTH" "$BASE/api/v1/admin/data-platform/sources/{name}/health"逐笔委托/成交/十档盘口由 wind_l2_import.py 从万得逐日 7z 手动导入
(schema/单位/坑见 [[quantdb-fields]] 第三章,含深市成交量≈2×、tick_data 单位混源):
python backend/scripts/wind_l2_import.py --archive /path/to/20260511.7z # 全市场
python backend/scripts/wind_l2_import.py --archive /path/to/20260511.7z --symbols 000001.SZ落盘 1_kline_data/l2_data/order_|trade_{code}_{date}.parquet + tick_data/{code}_{date}.parquet;
文件名即日期(20260511.7z → 20260511),增量跳过已存在,--force 覆盖,可断点续跑。
curl -s -H "$AUTH" "$BASE/api/v1/admin/data-platform/health-matrix?market=A"curl -s -H "$AUTH" "$BASE/api/v1/admin/data-platform/field-coverage"curl -s -H "$AUTH" "$BASE/api/v1/admin/data-platform/quality-alerts"通过 /api/v1/models/feature-catalog 获取。类别与特征数由 QuantDB l1_factors 动态生成(示例 version 20260831 返回 10 类 110 特征),随数据版本变化——以接口返回为准,不要硬编码类别清单。
推理研究涵盖:单日推理、批量多日推理、批量单日推理、推理历史、股票历史分数。
# 生成明日信号前置检查(确认数据就绪)
curl -s -H "$AUTH" "$BASE/api/v1/models/inference/precheck"# 对指定模型在指定日期执行推理(可能耗时数分钟)
curl -s -X POST "$BASE/api/v1/models/inference/run" -H "$AUTH" -H "$CT" \
-d '{"model_id":"xxx", "inference_date":"2026-08-07"}' \
-w "\nHTTP %{http_code}\n"批量推理支持两种模式,提交后立即返回 batch_id,逐日推理在后台执行:
A. 批量单日推理(range 模式)——区间内每个交易日逐日执行单日推理
curl -s -X POST "$BASE/api/v1/models/inference/batch" -H "$AUTH" -H "$CT" -d '{
"model_id": "xxx",
"mode": "range",
"start_date": "2026-08-01",
"end_date": "2026-08-07",
"top_k": 20,
"side": "both"
}'B. 批量多日推理(lookback 模式)——锚定日回溯 N 个交易日
curl -s -X POST "$BASE/api/v1/models/inference/batch" -H "$AUTH" -H "$CT" -d '{
"model_id": "xxx",
"mode": "lookback",
"anchor_date": "2026-08-07",
"window_days": 30, # 默认 = 模型 horizon,所有信号梯队仍持有中
"top_k": 20,
"side": "both",
"reuse_existing": true
}'参数完整说明:
| 参数 | 取值 | 说明 |
|---|---|---|
mode | range / lookback | range=日期区间逐日;lookback=锚定日回溯窗口 |
start_date / end_date | YYYY-MM-DD | range 模式必填,区间内逐日推理 |
anchor_date | YYYY-MM-DD | lookback 模式必填 |
window_days | 整数 | lookback 回溯天数(默认=模型 horizon) |
top_k | 整数 | 每日排名前 N 名 |
side | long / short / both | 多/空/双向 |
reuse_existing | 布尔 | 复用已存在的推理结果 |
concurrency | 整数 | 并发度 |
返回:HTTP 202 + batch_id。之后用 batch_id 轮询进度。
# 批量推理历史
curl -s -H "$AUTH" "$BASE/api/v1/models/inference/batches?page=1&page_size=20"
# 单个批次进度(status: pending/running/completed/failed)
curl -s -H "$AUTH" "$BASE/api/v1/models/inference/batch/{batch_id}"
# 删除批次记录
curl -s -X DELETE -H "$AUTH" "$BASE/api/v1/models/inference/batch/{batch_id}"/models/default 或 /models 选模型/inference/batch/{batch_id} 查进度,completed 后取结果# 推理历史(支持按 run_id/状态/日期过滤)
curl -s -H "$AUTH" "$BASE/api/v1/models/inference/runs?model_id=xxx&page=1&page_size=20"
# 单次推理结果明细(排名/信号/行业等)
curl -s -H "$AUTH" "$BASE/api/v1/models/inference/runs/{run_id}"
# 删除推理记录
curl -s -X DELETE -H "$AUTH" "$BASE/api/v1/models/inference/runs/{run_id}"# 某股票的历史推理分数趋势(用于交叉验证选股)
curl -s -H "$AUTH" "$BASE/api/v1/models/inference/stock/600036.SH/history?days=180"# 自动推理设置(每日定时)
curl -s -H "$AUTH" "$BASE/api/v1/models/inference/settings/{model_id}"
curl -s -X PUT -H "$AUTH" -H "$CT" "$BASE/api/v1/models/inference/settings/{model_id}" -d '{"auto_enabled": true}'
# 当前生效推理批次
curl -s -H "$AUTH" "$BASE/api/v1/models/inference/latest"# 某批次的聚合分析(per_symbol/groups/movers/daily/meta,含 IC/趋势/共识带)
curl -s -H "$AUTH" "$BASE/api/v1/models/inference/batch/{batch_id}/aggregate"融合模型(ensemble_config.json)本身无 pred.pkl,AI-IDE 回测/信号生成时会自动调用 generate_ensemble_pred:读取子模型 pred.pkl → 按 (datetime, instrument) 对齐 → 截面排名百分位加权融合 → 落到融合模型目录 pred.pkl。单模型无 pred 时提示"请先推理"。
curl -s -H "$AUTH" "$BASE/api/v1/news/sources"
# 返回: {sources: [{source_id, source_name, subscribe_url, type, folder_id, folder_name}], folders, total}curl -s -H "$AUTH" "$BASE/api/v1/news/articles" \
-G \
--data-urlencode "tickers=600519.SH,000858.SZ" \
--data-urlencode "industries=白酒,消费" \
--data-urlencode "sentiment=bullish" \
--data-urlencode "event_tags=财报,业绩预增" \
--data-urlencode "keyword=茅台" \
--data-urlencode "sort=sentiment_bullish" \
--data-urlencode "since=2026-08-01T00:00:00Z" \
--data-urlencode "page=1"过滤参数:
source_id / source_ids — 新闻源过滤folder_id — 文件夹过滤keyword — 标题关键词tickers — 股票代码(逗号分隔)industries — 行业sentiment — bullish / bearish / neutralevent_tags — 事件标签(财报/业绩预增/减持等)countries / regions — 国家/地区key_terms — 关键词(AI/半导体等)date_entities — 提及日期starred — 仅收藏strong_only — 仅强信号(|score|>=0.5)sort — time_desc(最新)/ time_asc / sentiment_bullish(利好强度)/ sentiment_bearish(利空强度)curl -s -H "$AUTH" "$BASE/api/v1/news/articles/{article_id}"curl -s -H "$AUTH" "$BASE/api/v1/news/enrichment/stats"
curl -s -X POST -H "$AUTH" "$BASE/api/v1/news/enrichment/run"
curl -s -X POST -H "$AUTH" "$BASE/api/v1/news/enrichment/rebuild-all"curl -s -X POST -H "$AUTH" "$BASE/api/v1/news/sources/{source_id}/refresh"当用户要求分析某股票/行业时,按此流程:
/news/articles 带 tickers + sentiment + since,看利好/利空/models/inference/stock/{symbol}/history 看历史推理分数趋势/admin/data-platform/health-matrix?market=A 确认数据完整/models/default 确认生效模型/admin/data-platform/daily-sync 提交增量同步feature-catalog 拿字段,再 run-training当用户要求挖掘新因子时,使用 [[rd-agent-factor-mining]] 技能(RD-Agent 自动演化管线)。 当用户要求按条件选股 / 筛选股票池时,使用 [[smart-strategy-stock-picking]] 技能(基于 QuantDB 字段字典的条件选股)。 当用户要求查询 QuantDB 数据 / 配置 API Key / 查看数据集字段时,使用 [[quantdb-sdk]] 技能。 当用户要求深度分析市场 / 数据挖掘 / 导出分析数据 / 生成投研报告时,使用 [[stock-market-analysis]] 技能。 当用户要求运行回测 / 对比策略 / 参数优化 / 分析回测结果时,使用 [[backtest-center]] 技能。 当用户要求用 AI 写策略 / 生成 Qlib 策略代码时,使用 [[ai-ide-strategy-writing]] 技能。 当用户要求模拟交易 / 下单 / 查持仓时,使用 [[simulation-trading]] 技能。 当用户要求分析批量推理结果 / 解读信号 / 选股决策 / 负分参考时,使用 [[batch-inference-analysis]] 技能。 当用户要求生成投研报告 / 深度研报 / 多Agent分析时,使用 [[stock-deep-research]] 技能。
注:投研报告由 [[stock-deep-research]] 技能(智能体自主版,任意大模型可跑)编排生成,落盘后由「技能中心 → 报告档案」统一浏览。
| 现象 | 排查 |
|---|---|
| 特征字典加载失败 | /models/feature-catalog 返回是否 200,看服务健康 |
| 数据匹配不到 | /admin/data-platform/health-matrix 看字段覆盖,/freshness 看新鲜度 |
| 新闻空白 | /news/sources 确认源存在,/news/enrichment/stats 看富化状态 |
| 训练失败 | /models/training-runs/{run_id} 查状态,看 features 是否在 parquet 中存在 |
© qusong0627, AGPL-3.0. Rendered from Markdown: HTML in the file is shown as text, images as links, and headings moved down two levels. Raw file
SKILL.md and 2 other files (references) in skills/quantmind-operations of qusong0627/QuantMind.
Open the folder on GitHubat commit 2e93d9a
Quantmind Operations next to the 5 skills that share the most tags, products or categories with it. Stars are the repository's; “used in” counts other GitHub owners with a copy.
| Skill | Stars | Used in | Tokens | Auto-check | Licence | Repo updated |
|---|---|---|---|---|---|---|
| Quantmind Operations this skillqusong0627/QuantMind | 1.7k | — | ~4k | Automated safety check: Pass | AGPL-3.0 | |
| MatplotlibzLanqing/codex-claude-academic-skills | 4.7k | 17 repos | ~2.9k | Automated safety check: Pass | MIT | |
| Exploratory Data Analysisspacering-net/codeg | 3.9k | 14 repos | ~3.6k | Automated safety check: Pass | MIT | |
| Scikit LearnzLanqing/codex-claude-academic-skills | 4.7k | 16 repos | ~3.9k | Automated safety check: Pass | BSD-3-Clause | |
| Chart Visualizationbytedance/deer-flow | 84k | 2 repos | ~840 | Automated safety check: Pass | MIT | |
| TimesFM Forecastinggoogle-research/timesfm | 34k | — | ~4.7k | Automated safety check: Pass | Apache-2.0 |
zLanqing/codex-claude-academic-skills
Low-level plotting library for full customization. An agent skill from zLanqing/codex-claude-academic-skills.
spacering-net/codeg
Perform comprehensive exploratory data analysis on scientific data files across 200+ file formats.
zLanqing/codex-claude-academic-skills
Machine learning in Python with scikit-learn. An agent skill from zLanqing/codex-claude-academic-skills.
bytedance/deer-flow
Picks a suitable chart type from 26 options for your data, maps the data to that chart's parameters and generates a chart image through a JavaScript script.
google-research/timesfm
Forecasts any univariate time series zero-shot with Google's TimesFM model, returning point forecasts and calibrated prediction intervals without training.
vercel/next.js
Benchmark React or Next.js changes on Vercel Sandbox VMs with paired A/B statistics: react PR/commit vs base, or Next.js PR/commit vs base, measured end-to-end through the bench/render-pipeline app…
qusong0627/QuantMind
Produces a post-market review report for the China A-share market from local QuantDB data, news sentiment and model signals, ending in a next-day direction call.
qusong0627/QuantMind
Queries Futu quotes, options, fundamentals and accounts and places orders through the Futu OpenAPI Python SDK, defaulting to simulated trading.
qusong0627/QuantMind
Turns a plain-language model training request into a validated QuantMind training config file that can be imported from the Model Training page.
qusong0627/QuantMind
Covers the Tiger Brokers OpenAPI Python SDK for market data, stock, futures and options trading, push subscriptions, a CLI and an MCP server, defaulting to paper trading.
qusong0627/QuantMind
Guides an agent through the Tiger Brokers OpenAPI C++ SDK for build setup, market data, orders and real-time push, defaulting to paper trading.
qusong0627/QuantMind
Guides building C# and .NET apps on the Tiger Brokers OpenAPI SDK: setup, market data, orders, accounts, options and real-time push, defaulting to paper trading.
Categories
QuantMind 平台运营操作技能 — 覆盖模型训练、模型管理、后台数据更新、字段信息查询、RSS 新闻对接与分析。在 QuantBot / Claude Code 中处理模型训练、数据同步、新闻分析等任务时使用。触发词:模型管理、数据更新、字段信息、RSS、新闻分析、查看数据、同步数据. Quantmind Operations is an agent skill from qusong0627/QuantMind.
Quantmind Operations fits situations like: data & Analytics work in your project.
Run `npx skills add qusong0627/QuantMind --skill quantmind-operations -a claude-code`. Or copy the skill folder (skills/quantmind-operations in qusong0627/QuantMind) into .claude/skills/quantmind-operations in your project. Claude Code loads it when a task matches its description.
Run `npx skills add qusong0627/QuantMind --skill quantmind-operations -a codex`. Or copy the skill folder (skills/quantmind-operations in qusong0627/QuantMind) into .agents/skills/quantmind-operations in your project. Codex loads it when a task matches its description.
Cursor, Gemini CLI, GitHub Copilot and OpenCode also load SKILL.md folders. With the skills CLI, run `npx skills add qusong0627/QuantMind --skill quantmind-operations -a cursor` (or -a gemini-cli, github-copilot or opencode for the others). To copy it by hand, put the folder in .cursor/skills/quantmind-operations, .gemini/skills/quantmind-operations, .github/skills/quantmind-operations and .opencode/skills/quantmind-operations in your project.
Going by SKILL.md and its folder, Quantmind Operations needs the command-line tools its instructions call (curl, python and python3). Our summary lists: Python 3.
SKILL.md contains no URLs. Its commands use curl, which can reach the network depending on how they are called. This is read from the text; nothing was executed.
Our automated static check of SKILL.md found no risky patterns, such as piping downloads into a shell, reading credential files or hidden Unicode. It is not a guarantee. Review the folder before installing.
Quantmind Operations is published under the AGPL-3.0 licence (the repository's licence). It allows redistribution, so the full SKILL.md is shown on this page.
About 4k tokens (SKILL.md is roughly 16k characters). Agents keep only the skill's name and description in context until a task matches; then they load SKILL.md in full. Its references folder adds about 2k tokens, read only when the agent opens those files.
Skills that share tags, products or a category with Quantmind Operations: Matplotlib (zLanqing/codex-claude-academic-skills, 4.7k stars), Exploratory Data Analysis (spacering-net/codeg, 3.9k stars), Scikit Learn (zLanqing/codex-claude-academic-skills, 4.7k stars) and Chart Visualization (bytedance/deer-flow, 84k stars). The comparison table on this page puts their stars, adoption, token cost, safety result and licence side by side.
qusong0627 (a GitHub user) maintains it in qusong0627/QuantMind, which has 1,723 GitHub stars. The repository holds 27 skills in this directory. The repository was last updated on October 9, 2026.
Source: qusong0627/QuantMind on GitHub. Facts on this page come from the repository at the commit we read; the author's words are quoted as theirs.