← 返回 Skills 市场

a-share-portfolio-calibrator

Name: a-share-portfolio-calibrator
Author: minguncle

作者 minguncle · GitHub ↗ · v1.0.0 · MIT-0

cross-platform ✓ 安全检测通过

总下载

当前安装

版本数

在 OpenClaw 中安装

/install a-share-portfolio-calibrator

功能描述

把持仓截图或持仓列表标准化成组合工作包，输出仓位、权重、集中度、主题暴露和执行约束。适用于用户发账户截图、贴持仓表、想先诊断组合形态而不是立刻下结论的场景。

使用说明 (SKILL.md)

A Share Portfolio Calibrator

Overview

这个 skill 专门把“持仓信息”整理成后续可路由的标准格式。它先做提取和校准，不跳步到买卖动作，也不假定之前线程里的组合数据仍然有效。

Hard Rules

这是组合诊断 skill，不是直接下指令的 skill。
用户给截图时，优先提取能确认的字段；看不清就标缺口，不乱猜。
必须计算权重、集中度、主题暴露和可用仓位。
如果没有当日市场底稿，不能把组合诊断直接升级成最终调仓建议。
不把“有票在涨”误判为“组合健康”。

When To Use

用户贴了账户截图或持仓列表
需要先看仓位结构是否失衡
用户想知道自己到底在押什么风格
后续要做留减清或新建仓，但还没整理组合底稿

Workflow

Step 1: 标准化账户层信息

优先抽取：

总资产
股票市值
可用现金
可取资金
是否存在冻结或不可卖仓位

Step 2: 标准化持仓层信息

对每只票提取：

代码与名称
持股数量
可用数量
成本价
现价
盈亏金额或盈亏比
市值

Step 3: 计算组合指标

至少计算：

单票权重
前三大持仓占比
现金比
同主题持仓合计
风格暴露，例如科技、资源、红利、周期、成长

Step 4: 识别隐性约束

例如：

单票过度集中
可卖仓位不足
多票本质上押的是同一条主线
账面浮盈过大导致回撤容忍度偏低
账面浮亏票占用决策带宽

模板见 calibration-template.md。

Step 5: 输出组合工作包

输出给后续 skill 的不是一堆原始数据，而是一份清晰的组合 packet：

组合结构
风格集中点
可调整资源
主要风险点
待确认缺口

Tool Rules

截图优先用图像能力或低风险桌面读取
若用户提供文字列表，直接结构化处理，不额外重复提问
组合权重计算优先使用账户总市值或总资产两个口径都给，避免混淆

Output Contract

默认输出：

账户快照
标准化持仓表
组合集中度
主题与风格暴露
主要约束
待补信息

Practical Notes

若用户仓位很多，不要全量长篇点评，先抓最大权重和重复暴露
若截图显示某只票可用仓位远小于持仓仓位，要提示 T+1 或冻结影响
完成后若用户问“怎么调”，下一步交给动作路由

安全使用建议

This skill appears coherent and only standardizes user-provided holdings into a diagnostic packet. Before using it: (1) verify any screenshots you provide do not contain passwords, API keys, account logins, or other sensitive identifiers; (2) confirm the agent will only read the images/text you explicitly upload or paste (don’t allow it to scan arbitrary directories); (3) understand the skill explicitly states it won’t place trades—treatment/actuation should be a separate, deliberate step; (4) review the standardized output for correctness before taking action. If you need the agent to access local screenshots, make sure you trust the environment and confirm exactly which files will be read.

功能分析

Type: OpenClaw Skill Name: a-share-portfolio-calibrator Version: 1.0.0 The skill bundle is designed to standardize and analyze A-share stock portfolio data from user-provided screenshots or text lists. The logic in SKILL.md and the template in calibration-template.md focus entirely on extracting financial metrics (e.g., weights, concentration, and theme exposure) without any evidence of data exfiltration, malicious command execution, or harmful prompt injection. The requested access to 'desktop reading' is consistent with the stated purpose of processing screenshots within the OpenClaw environment.

能力评估

✓ Purpose & Capability

Name, description, and runtime instructions all describe the same task (extract holdings from screenshots/text, compute weights/exposures, produce a calibration packet). There are no unrelated required env vars, binaries, or config paths.

ℹ Instruction Scope

SKILL.md limits itself to extraction, computation, and producing a standard packet. It allows using image capability or '低风险桌面读取' for screenshots — which is reasonable for this purpose but means the agent may read user-supplied images or local screenshots. Recommend confirming with users before accessing any local files or sensitive images.

✓ Install Mechanism

No install spec and no code files; the skill is instruction-only so it does not download or install artifacts to the host system.

✓ Credentials

The skill requests no environment variables, credentials, or config paths. Its declared needs are proportional to its stated functionality.

✓ Persistence & Privilege

always is false and there is no install or persistent background behavior. The skill does not request elevated or persistent privileges.

如何使用

确保已安装 OpenClaw（本地或 Docker 部署）
在对话框中输入安装命令：/install a-share-portfolio-calibrator
安装完成后，直接呼叫该 Skill 的名称或使用 /a-share-portfolio-calibrator 触发
根据 Skill 的参数说明提供必要输入，即可获得结构化输出

版本历史

v1.0.0

A-share Portfolio Calibrator launches with essential features for standardizing and analyzing portfolio snapshots. - Extracts and normalizes account and position data from screenshots or text lists. - Calculates key metrics: weights, concentration, thematic exposure, and actionable constraints. - Flags missing or uncertain data fields without guessing. - Outputs a structured portfolio packet for further analysis, not direct trading. - Clearly separates diagnosis from actionable recommendations.

元数据

Slug a-share-portfolio-calibrator

版本 1.0.0

许可证 MIT-0

累计安装 0

当前安装数 0

历史版本数 1

常见问题