← 返回 Skills 市场
TVFetch
作者
Tarun Khatri
· GitHub ↗
· v1.0.0
· MIT-0
94
总下载
0
收藏
0
当前安装
1
版本数
在 OpenClaw 中安装
/install tvfetch
功能描述
Fetch TradingView market data — historical OHLCV bars, live price streams, symbol search, technical indicators, and statistical analysis for any symbol (stoc...
安全使用建议
This skill appears to be what it claims: a reverse-engineered TradingView fetcher with helpful CLI scripts. Before installing or running: 1) Review the included Python code if you can, or run in an isolated virtualenv/container to avoid contaminating your system Python. 2) Be cautious with the optional TV_AUTH_TOKEN — the skill advises copying a JWT from your browser; treat that token like a password and only save it if you trust the code and environment (it will be stored in ~/.tvfetch/.env or keyring). 3) The skill may install dependencies (websocket-client, httpx, pandas, etc.) if you choose to pip install; prefer virtualenv. 4) If you prefer not to expose your TradingView account token, use anonymous mode (default) understanding intraday bar limits. 5) If you want extra assurance, run the scripts in mock mode or inspect network connections the library makes (it targets data.tradingview.com), and avoid running pip install -e unless you reviewed the repository.
功能分析
Type: OpenClaw Skill
Name: tvfetch
Version: 1.0.0
The tvfetch skill is a comprehensive tool for retrieving market data from TradingView via a reverse-engineered WebSocket protocol. It features historical data fetching, live streaming, technical indicators (indicators.py), and statistical analysis (analyze.py). The skill manages sensitive authentication tokens locally in ~/.tvfetch/.env and includes a deployment script (scripts/sync.sh) for multi-agent environments. All behaviors, including network requests to legitimate TradingView endpoints (data.tradingview.com) and local file management for caching, are clearly aligned with the stated purpose and lack evidence of malicious intent, data exfiltration, or unauthorized execution.
能力标签
能力评估
Purpose & Capability
Name/description (TradingView OHLCV, streaming, indicators) lines up with included Python library and CLI scripts (fetch.py, stream.py, search.py, indicators, analyzers). Optional auth token, cache, and fallback behavior are plausible for this functionality.
Instruction Scope
SKILL.md directs the agent to run bundled Python scripts, check and write config under ~/.tvfetch, and optionally install the package from the skill directory (pip install -e ${CLAUDE_SKILL_DIR}). It also instructs how to obtain a TradingView auth_token via browser DevTools (document.cookie) which is sensitive but logically tied to increasing intraday limits. The instructions read/write only the skill's config paths and .env files (and may consult keyring), which is within scope for this skill.
Install Mechanism
There is no platform install spec, but SKILL.md recommends pip install -e from the skill directory to enable the library. Installing local Python code is expected to enable the CLI and modules, but note that pip installing editable code will write/execute package metadata on the host — review code before installing or install in an isolated environment.
Credentials
The skill does not require environment variables by default. It supports an optional TV_AUTH_TOKEN (CLI/ENV/.env/keyring) to increase bar limits, and optional TVFETCH_CACHE_PATH/TVFETCH_PROXY. Access to a single service token and local cache path is proportional to the stated functionality. The config will check keyring and .env files for a token — this is reasonable but users should be aware any token stored is a TradingView auth cookie (sensitive).
Persistence & Privilege
always:false and user-invocable:true. A SessionStart hook runs a non-blocking check-config.sh which prints warnings and may create ~/.tvfetch — this is limited scope and does not alter other skills or system-wide settings. The skill writes/reads only its own config/cache paths.
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 tvfetch - 安装完成后，直接呼叫该 Skill 的名称或使用
/tvfetch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
Initial release of tvfetch — fetch and analyze market data from TradingView with no API key.
- Fetch historical OHLCV bars, live price streams, search symbols, and compute key technical indicators (RSI, MACD, SMA, EMA, Bollinger Bands, etc.).
- Supports stocks, crypto, forex, futures, indices, and commodities for any TradingView symbol.
- Automatically interprets user intent from natural language and resolves symbols to exchange-prefixed codes.
- Handles anonymous usage limits and guides users to set up authentication for higher data limits.
- Outputs can be shown as tables or downloaded in CSV, JSON, or Parquet formats.
元数据
常见问题
TVFetch 是什么？
Fetch TradingView market data — historical OHLCV bars, live price streams, symbol search, technical indicators, and statistical analysis for any symbol (stoc... 它是一个面向 Claude Code / OpenClaw 的 AI Agent Skill 插件，目前累计下载 94 次。
如何安装 TVFetch？
在 OpenClaw 或 Claude Code 对话框中运行命令「/install tvfetch」即可一键安装，无需额外配置。
TVFetch 是免费的吗？
是的，TVFetch 完全免费，采用 MIT-0 许可证，可自由下载、安装和使用。
TVFetch 支持哪些平台？
TVFetch 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 TVFetch？
由 Tarun Khatri（@tarun-khatri）开发并维护，当前版本 v1.0.0。
推荐 Skills