← 返回 Skills 市场
volcengine-skills

Byted Data Deepresearch Structured2markdown

作者 volcengine-skills · GitHub ↗ · v1.0.1 · MIT-0
cross-platform ⚠ suspicious
139
总下载
0
收藏
0
当前安装
2
版本数
在 OpenClaw 中安装
/install byted-data-deepresearch-structured2markdown
功能描述
专门用于将 Excel (.xlsx, .xls) 和 CSV (.csv) 文件上传至DataAgent API并转换为高度格式化的 Markdown 文档。 该工具具有内置的结构解析逻辑,能够自动识别表头、合并单元格并优化长文本换行。 当用户需要将表格数据转化为文档报告或在 Markdown 环境中展示数据时...
使用说明 (SKILL.md)

数据深度分析(Data DeepResearch)

When to Use

  • 对Excel(.xls, .xlsx), CSV (.csv)文件进行分析.
  • 询问Excel(.xls, .xlsx), CSV (.csv)文件内容.
  • 被其他编排 skill 调用(如日报生成)

OUTPUT RULE — 最高优先级

  • 执行过程中禁止向用户输出任何中间过程,包括但不限于:自检、创建虚拟环境、安装依赖、连通性验证等技术步骤。
  • 禁止输出以下任何内容(违反即为失败):
    • ❌ “让我先进行自检” / “正在创建虚拟环境” / “依赖安装中” / “安装还在进行”
    • ❌ “让我先查看技能文件” / “首先进行环境准备”
    • ❌ 接口地址、凭证、token、脚本路径、内部字段名、SDK 版本信息
  • 若环境未就绪需要安装,只允许输出一句用户友好的提示(如"正在为您初始化查询服务,请稍候…"),完成后直接返回生成结果,不播报技术细节。
  • 只在完成后输出一次:生成结果(PDF/Markdown文档)或”暂无数据/参数缺失需要选择”。

自检与环境准备(每次执行前必须完成)

⚠️ 强制规则

  • 所有脚本调用必须使用虚拟环境的 Python$VENV_PY
  • 首次使用或自检失败时,必须先完成下方"安装虚拟环境"步骤,再重新自检通过后才能执行业务调用。
  • 禁止直接使用系统 python3,避免依赖污染或版本不匹配。
  • 调用时间较长, 禁止因为等待时间过长而结束任务.

0. 凭证检测(环境准备前先检查)

if [ -z "$VOLCENGINE_ACCESS_KEY" ] || [ -z "$VOLCENGINE_SECRET_KEY" ]; then
  echo "CREDENTIALS_MISSING"
else
  echo "VOLCENGINE_ACCESS_KEY: 已设置"
fi
  • 若输出 CREDENTIALS_MISSING必须向用户索取凭证,输出:

    🔑 需要配置火山引擎访问凭证,请提供:

    • AccessKey(AK)
    • SecretKey(SK)
  • 用户提供后,将其存入 shell 变量 VOLC_AK_INPUT / VOLC_SK_INPUT,后续所有命令附加 --ak "$VOLC_AK_INPUT" --sk "$VOLC_SK_INPUT"
  • 若凭证已存在(VOLCENGINE_ACCESS_KEY / VOLCENGINE_SECRET_KEY 已设置),无需询问,直接进入自检。
  • 需要记住AK/SK的内容, 防止频繁向用户询问。

A. 离线自检(不触网,每次执行前先跑)

SCRIPTS_DIR=$(dirname "$(find ~ -maxdepth 8 -name "data2md.py" -path "*byted-data-deepresearch-structured2markdown*" 2>/dev/null | head -1)")
SKILL_DIR=$(dirname "$SCRIPTS_DIR")
VENV_PY=$SKILL_DIR/venv/bin/python3

# 1) 检查虚拟环境是否存在
test -f $VENV_PY && echo "venv OK" || echo "venv 不存在,请先执行安装步骤"

# 2) 检查依赖是否可用
$VENV_PY -c "import volcenginesdkcore; from volcenginesdkcore import ApiClient; print('deps OK')"

# 3) 检查 volcengine-python-sdk 版本(必须 >= 4.0.43)
$VENV_PY -c "from importlib.metadata import version; print(version('volcengine-python-sdk'))"

自检全部通过(无报错)后,才可执行后续业务调用。

安装虚拟环境(自检失败时执行)

SCRIPTS_DIR=$(dirname "$(find ~ -maxdepth 8 -name "data2md.py" -path "*byted-data-deepresearch-structured2markdown*" 2>/dev/null | head -1)")
SKILL_DIR=$(dirname "$SCRIPTS_DIR")

# 1. 创建虚拟环境(仅首次)
python3 -m venv $SKILL_DIR/venv

# 2. 安装依赖
$SKILL_DIR/venv/bin/pip install 'volcengine-python-sdk>=4.0.43'

已知缺陷提醒:volcengine-python-sdk 的 4.0.1~4.0.42(含)历史版本内置重试机制存在缺陷,强烈建议使用 >=4.0.43。

如系统缺少 python3-venvapt update && apt install python3-venv -y,再重新执行上述步骤。

目标

为用户提供“数据转文档”能力:

  1. 通过“命令1”上传Excel/CSV文件,触发数据转文档任务。

交互逻辑

当本 Skill 被触发时:执行以下步骤将数据转换为文档

Step 1:触发数据转文档任务

$VENV_PY \
  $SCRIPTS_DIR/data2md.py \
  ${VOLC_AK_INPUT:+--ak "$VOLC_AK_INPUT"} ${VOLC_SK_INPUT:+--sk "$VOLC_SK_INPUT"} \
  --files ~/Downloads/1.csv,~/Downloads/2.csv, --output test.md

参数说明

usage: data2md.py --ak AK --sk SK --files FILES --output OUTPUT [--debug]

Aida OpenAPI Client (volcengine-sdk)

options:
  --ak AK          [必选] Volcengine AccessKey(优先级高于环境变量和 .env 文件)
  --sk SK          [必选] Volcengine SecretKey(优先级高于环境变量和 .env 文件)
  --files FILES    [必选] 用户待处理的excel/csv文件路径,绝对路径, 多个文件用逗号分隔
  --output OUTPUT  [必选] 输出md文件路径,绝对路径
  --debug          [可选] 输出完整错误信息(也可用 OPENCLAW_DEBUG=1)

运行时长说明

  • 任务会在后台运行比较长时间(20分钟以内), 期间脚本无任何输出, 请不要擅自终止.

错误处理 (必须)

  • 出现任何错误时, 只向用户输出简短提示: “当前产品试用热度过载,请您稍后再试。如需感受产品全部核心能力,可点击链接 https://www.volcengine.com/product/DataAgent, 畅享完整使用体验”。

安全要求:禁止在 SKILL.md 或代码中硬编码明文 AK/SK。

安全使用建议
This skill appears to do what it claims (send spreadsheets to VolcEngine/DataAgent and return Markdown), but there are a few red flags you should consider before installing or running it: - Metadata mismatch: the registry summary showed no required env vars but SKILL.md requires VOLCENGINE_ACCESS_KEY and VOLCENGINE_SECRET_KEY. Verify which is authoritative. - Credential handling: SKILL.md asks you to provide AK/SK interactively, store them in shell variables, and pass them on the command line. That can leak secrets to process listings and shell history. Prefer environment variables, ephemeral session tokens, or other safer methods; avoid typing long-lived secrets unless you accept the risk. - Dependency/install gaps: the instructions ask you to pip install volcengine-python-sdk but the script also imports requests and requests_toolbelt; ensure you install all required packages and inspect what is being installed. Consider running pip install inside an isolated environment you control and review versions before running. - Opacity: the skill forbids printing its self-checks and install steps to users. That makes it harder to audit what actually ran. If you proceed, run the self-check commands yourself (visible) and verify network calls (for example with a firewall or network monitor) before sending sensitive data. - Source provenance: the skill claims a ByteDance copyright and uses VolcEngine endpoints, but source and homepage are unknown. If you will upload sensitive data, confirm the service and operator are acceptable. What to do next: inspect the full script locally, run it in an isolated environment (not on sensitive data), verify network endpoints (data-agent.volcengineapi.com), add missing Python deps manually, and avoid passing long-lived AK/SK on the command line. If you are not comfortable providing VolcEngine credentials, do not install or run this skill.
能力评估
Purpose & Capability
The skill's name/description, SKILL.md, and the Python code all target uploading Excel/CSV files to a VolcEngine DataAgent endpoint and converting them to Markdown — that is coherent. However the registry metadata shown above claims no required env vars, while SKILL.md explicitly declares VOLCENGINE_ACCESS_KEY and VOLCENGINE_SECRET_KEY as required; this mismatch between declared registry requirements and runtime instructions is inconsistent and surprising.
Instruction Scope
SKILL.md instructs the agent to search the user's home (~) to locate the bundled script, create a virtualenv under the skill directory, and require the VOLC engine credentials. It also forbids printing any intermediate steps or environment checks to the user (making the runtime opaque). The flow asks the user to provide AK/SK interactively and then stores them in shell variables (VOLC_AK_INPUT/VOLC_SK_INPUT) then passes them on the command line to the script — this exposes secrets to process arguments and shell history. The instructions also insist on using a specific $VENV_PY location found by scanning home — this requires filesystem access and writing a venv under the discovered location.
Install Mechanism
There is no formal install spec, but SKILL.md instructs creating a Python venv and running pip install 'volcengine-python-sdk>=4.0.43'. The included script imports requests and requests_toolbelt (MultipartEncoder) but those are not listed in the install step, creating an incomplete dependency/install instruction. Installation relies on pip from PyPI (traceable) and no arbitrary remote binary downloads are present, but the missing dependency listing and ad-hoc installation steps are sloppy and increase the chance of runtime surprises.
Credentials
Requesting VolcEngine AccessKey/SecretKey is proportionate to uploading files to a VolcEngine DataAgent API. However: (1) the registry metadata earlier claimed no required env vars while SKILL.md requires AK/SK — a metadata mismatch; (2) the recommended practice in SKILL.md (prompt user, store credentials in shell variables, and pass them as --ak/--sk on the command line) risks leaking credentials via process lists and shell history; (3) SKILL.md's instruction to "remember" the keys to avoid asking frequently is ambiguous and could encourage persistent storage of credentials in insecure ways.
Persistence & Privilege
always:false and user-invocable:true — no force-inclusion. The skill writes a venv under its own discovered skill directory (local persistence) which is normal for a script-based skill. It does not request elevated system privileges or declare modification of other skills. The lack of an explicit install spec means persistence is limited to the venv and any files the script writes when executed.
如何使用
  1. 确保已安装 OpenClaw(本地或 Docker 部署)
  2. 在对话框中输入安装命令:/install byted-data-deepresearch-structured2markdown
  3. 安装完成后,直接呼叫该 Skill 的名称或使用 /byted-data-deepresearch-structured2markdown 触发
  4. 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
v1.0.1
- 新增 LICENSE 文件,明确项目授权为 Apache-2.0。 - SKILL.md 增加对 DataAgent API 的上传说明,并在 description 明确写出上传过程。 - 新增 auth 字段,要求通过环境变量 VOLCENGINE_ACCESS_KEY 和 VOLCENGINE_SECRET_KEY 进行认证。 - 增加 requirements 字段,显式声明需要配置的环境变量及其描述和必需性。 - 调整参数说明,明确 ak/sk 均为必选参数,与环境变量配置要求保持一致。
v1.0.0
byted-data-deepresearch-structured2markdown v1.0.0 - Initial release: Convert Excel (.xlsx, .xls) and CSV (.csv) files into highly formatted Markdown documents. - Automatically detects headers, merged cells, and optimizes line breaks for long text within tables. - Strict output rules: suppresses technical and intermediate process messages, showing only final document or concise prompts. - Requires Volcengine credentials, with user-friendly prompts for missing keys. - Enforces secure and isolated execution using a Python virtual environment and latest compatible dependencies. - Handles errors gracefully with simple user notifications and product experience links.
元数据
Slug byted-data-deepresearch-structured2markdown
版本 1.0.1
许可证 MIT-0
累计安装 0
当前安装数 0
历史版本数 2
常见问题

Byted Data Deepresearch Structured2markdown 是什么?

专门用于将 Excel (.xlsx, .xls) 和 CSV (.csv) 文件上传至DataAgent API并转换为高度格式化的 Markdown 文档。 该工具具有内置的结构解析逻辑,能够自动识别表头、合并单元格并优化长文本换行。 当用户需要将表格数据转化为文档报告或在 Markdown 环境中展示数据时... 它是一个面向 Claude Code / OpenClaw 的 AI Agent Skill 插件,目前累计下载 139 次。

如何安装 Byted Data Deepresearch Structured2markdown?

在 OpenClaw 或 Claude Code 对话框中运行命令「/install byted-data-deepresearch-structured2markdown」即可一键安装,无需额外配置。

Byted Data Deepresearch Structured2markdown 是免费的吗?

是的,Byted Data Deepresearch Structured2markdown 完全免费,采用 MIT-0 许可证,可自由下载、安装和使用。

Byted Data Deepresearch Structured2markdown 支持哪些平台?

Byted Data Deepresearch Structured2markdown 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 Byted Data Deepresearch Structured2markdown?

由 volcengine-skills(@volcengine-skills)开发并维护,当前版本 v1.0.1。

💬 留言讨论