← 返回 Skills 市场
2289
总下载
4
收藏
33
当前安装
3
版本数
在 OpenClaw 中安装
/install daily-stock-analysis
功能描述
Deterministic daily stock analysis skill for global equities. Use when users need daily analysis, next-trading-day close prediction, prior forecast review, r...
安全使用建议
Install only if you are comfortable with the skill saving stock reports in your chosen working directory and reusing those reports in later runs. Use a dedicated project folder rather than a sensitive home or business-data directory, review any proposed legacy-file migration before approving it, and treat the financial output as research rather than investment advice.
功能分析
Type: OpenClaw Skill
Name: daily-stock-analysis
Version: 1.0.2
The skill bundle is classified as benign. All Python scripts (`_report_utils.py`, `calc_accuracy.py`, `report_manager.py`) strictly adhere to the `working_directory` constraint, preventing file operations outside the designated scope. They explicitly avoid following symlinks and perform non-destructive file migrations. The `SKILL.md` and `references/security.md` files contain clear instructions that reinforce these security boundaries for the AI agent, with no evidence of prompt injection attempts to subvert the agent's behavior or exfiltrate data. The scripts do not perform any network calls, aligning with the stated security rules, and there are no signs of credential theft, backdoors, or other malicious intent.
能力评估
Purpose & Capability
The stated stock-analysis purpose matches the artifacts: report generation, prior forecast review, rolling accuracy calculation, public financial source guidance, and local markdown output all support the workflow.
Instruction Scope
The skill instructs file reads and writes, but scopes them to working_directory, caps report history reads, limits report filenames, and requires explicit user confirmation before migrating legacy files.
Install Mechanism
There is no installer, package install, startup hook, or background setup; the bundled Python scripts are invoked on demand.
Credentials
Local report storage and public market-data lookups are proportionate for daily stock analysis; the artifacts do not request credentials, private accounts, or broad system access beyond the user-chosen working directory.
Persistence & Privilege
The skill persists report files for future review and recommends a user-created recurring task, but it does not create its own scheduler, daemon, elevated privilege path, or hidden persistence.
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 daily-stock-analysis - 安装完成后，直接呼叫该 Skill 的名称或使用
/daily-stock-analysis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
v1.0.2
Version 1.0.2 – Major update with new file structure, deterministic output rules, and Python script integration.
- Enforces working directory and canonical report/output location for better file management.
- Integrates required Python scripts for report planning and accuracy calculation.
- Adds compatibility/minimal modes for environments without Python or with limited resources.
- Introduces stricter overwrite/versioning logic for daily report files.
- Adds new reference documents: minimal_mode.md, security.md, sources.md.
- Streamlines required and optional output fields; mandates actionable self-improvement in every run.
v1.0.1
- Introduced a structured self-evolution mechanism: each analysis run now generates explicit learning records that adjust future assumptions and factor weights based on observed forecast errors.
- Added a feedback loop for continuous improvement, including root-cause tracking, action recording, and validation/rollback of learnings.
- Expanded `improvement_actions` to support learning notes, scope (ticker-specific or cross-ticker), and expiry conditions.
- Updated documentation to clarify the looped, learn-from-history approach rather than one-off reporting.
- No changes to core operation modes, report structure, or compliance disclaimer.
v1.0.0
Initial release with daily stock analysis and forecasting for major global markets.
- Provides next-trading-day close price predictions with confidence levels and assumptions.
- Includes trading recommendations (Buy/Hold/Sell/Watch) with rationale.
- Reviews previous forecasts against actual results, tracks error metrics, and explains misses/hits.
- Computes rolling forecast accuracy (1d/3d/7d/30d/custom windows).
- Implements continuous-improvement loop using postmortem analysis.
- Automatically saves daily analysis reports as markdown files for history and quality tracking.
元数据
常见问题
Daily Stock Analysis 是什么？
Deterministic daily stock analysis skill for global equities. Use when users need daily analysis, next-trading-day close prediction, prior forecast review, r... 它是一个面向 Claude Code / OpenClaw 的 AI Agent Skill 插件，目前累计下载 2289 次。
如何安装 Daily Stock Analysis？
在 OpenClaw 或 Claude Code 对话框中运行命令「/install daily-stock-analysis」即可一键安装，无需额外配置。
Daily Stock Analysis 是免费的吗？
是的，Daily Stock Analysis 完全免费（开源免费），可自由下载、安装和使用。
Daily Stock Analysis 支持哪些平台？
Daily Stock Analysis 跨平台运行，可在任意部署了 OpenClaw / Claude Code 的环境中使用（cross-platform）。
谁开发了 Daily Stock Analysis？
由 Xavi He（@hexavi8）开发并维护，当前版本 v1.0.2。
推荐 Skills