📦 Data Profiler — 数据画像
v2.1.0一键解析数据分布、质量指标与模式，为ETL与质量评估提供全景视图。
0· 1.1k·1 当前·1 累计
by 下载技能包
最后更新
2026/2/26
安全扫描
OpenClaw
安全
medium confidenceThe skill's requirements and instructions are consistent with a data-profiling tool: it needs python3, processes user-supplied datasets, and does not request unrelated credentials or external installs.
评估建议
This skill appears coherent for profiling construction datasets, but check the following before installing: 1) Confirm the runtime environment has python3 plus pandas/numpy (the skill assumes those packages). 2) Avoid providing sensitive production data unless you trust the environment—the skill expects to read files from the filesystem, and claw.json requests filesystem permission. 3) Verify there are no hidden network behaviors at runtime (SKILL.md shows only local profiling code in the visibl...详细分析 ▾
✓ 用途与能力
Name/description align with what the skill asks for: it profiles construction data and only requires python3. The documented code (pandas/numpy-based profiling) matches the stated purpose. There are no unrelated env vars, cloud credentials, or unexpected binaries requested.
ℹ 指令范围
SKILL.md provides concrete profiling code and instructs the agent to accept user-provided CSV/Excel/JSON or file paths and to validate inputs. This stays within the profiling scope. Note: the skill assumes the agent will read dataset files from the filesystem (user-supplied or provided paths); ensure files contain only data you intend to share.
✓ 安装机制
There is no install spec and no downloads—this is instruction-only. That lowers risk. Minor friction: the code uses pandas/numpy but no install instructions are included, so the runtime must already provide these Python packages.
✓ 凭证需求
No environment variables, credentials, or external tokens are requested. The lack of secrets is proportional to the stated functionality.
ℹ 持久化与权限
always:false and no autonomous-disable flags—behavior is normal. claw.json declares a 'filesystem' permission, which is reasonable for a skill that reads user files, but you should confirm the runtime's meaning of this permission (e.g., whether it grants broad file access beyond user-provided paths).
安全有层次，运行前请审查代码。
运行时依赖
🖥️ OSmacOS · Linux · Windows
版本
latestv2.1.02026/2/7
**Summary: Enhanced profiling and construction-specific insights** - Improved profiling to capture characteristics, distributions, nulls, quality metrics, and patterns in construction data. - Added construction-specific pattern and column type detection (e.g., csi_code, project_id, cost fields). - Now includes quality issue flags and actionable recommendations. - Reports memory usage, duplicate rows, and overall quality score. - Documentation updated with clearer business case, technical overview, and usage examples.
● 无害
安装命令
点击复制官方npx clawhub@latest install data-profiler
镜像加速npx clawhub@latest install data-profiler --registry https://cn.longxiaskill.com