技能 deeptools
📦

deeptools

内容修订 r1 高风险 ⚙️ 外部命令📁 文件系统访问

使用 deepTools 分析 NGS 数据

NGS 分析通常需要谨慎选择命令、进行归一化并验证文件。此技能帮助用户构建用于覆盖度、质量控制（QC）、比较和可视化的 deepTools 工作流。

支持: Claude Codex Code(CC)
⚠️ 38 差

使用我的智能体安装

将此请求复制给你的智能体，其中包含权威 Skill 页面和清单。

智能体请求
Review the Skillstore skill "deeptools" from https://skillstore.io/skills/k-dense-ai-deeptools.md and its manifest at https://skillstore.io/api/skills/k-dense-ai-deeptools/manifest. Verify the artifact. Stop and obtain explicit user consent before installing or changing files.

你的智能体仍应展示其计划，并按安全策略请求必要确认。

Agent 可读资源

当 AI Agent、爬虫或脚本需要干净上下文，而不是读取完整页面时，请使用这些链接。

Markdown 详情 GET /skills/k-dense-ai-deeptools.md 已签名 manifest GET /api/skills/k-dense-ai-deeptools/manifest 已签名 lockfile GET /api/skills/k-dense-ai-deeptools/lockfile

测试它

正在使用“deeptools”。 我需要比较 treatment 和 control ChIP-seq BAM 文件。

预期结果:

推荐的 bamCompare 工作流，包含 log2 ratio 输出、reads 计数缩放、重复 reads 处理以及后续热图指导。

正在使用“deeptools”。 我的 BAM 文件已经准备好了，但我不知道该运行哪些质量控制图。

预期结果:

一份质量控制计划，涵盖样本相关性、PCA、覆盖度分布、片段大小检查和 ChIP 富集 fingerprints。

正在使用“deeptools”。 我想要链特异性 RNA-seq 覆盖度轨道。

预期结果:

使用 bamCoverage 的覆盖度计划，包含链过滤、CPM 归一化，以及不要延伸 RNA-seq reads 的警告。

安全审计

高风险
v7 • 2026/7/5 打开版本化报告

Most SKILL.md and reference findings are false positives caused by Markdown command examples and genomics option names. The confirmed risks are in scripts/workflow_generator.py, which writes shell scripts with unescaped user-provided paths and writes to caller-selected output files. No prompt injection or data exfiltration intent was found.

8
已扫描文件
2,778
分析行数
8
复核项
0
已忽略误报

已确认安全问题 (1)

高
scripts/workflow_generator.py:40-44scripts/workflow_generator.py:115-121scripts/workflow_generator.py:283-287
Generated Workflow Scripts Use Unescaped Shell Inputs
workflow_generator.py places CLI parameters such as BAM paths, BED paths, and output directories directly into bash templates. A crafted value can break quoting and run unintended shell commands when the generated script is executed.
The code assigns command-line values directly inside generated shell syntax and later uses those variables in commands. Exploitation requires running the generated workflow, so confidence is high but not absolute.
能力复核项 (8)

这些是真实的本地能力，对此技能可能是预期行为，因此需要复核，但不会按已确认恶意行为计分。

中
scripts/workflow_generator.py:35
Unix shell invocation
script = f"""#!/bin/bash
This starts a generated bash workflow template that embeds caller-provided paths and settings into shell syntax without shell escaping. The generator does not execute it immediately, but the produced script can run unintended commands if executed with crafted values.
中
scripts/workflow_generator.py:110
Unix shell invocation
script = f"""#!/bin/bash
This starts a generated bash workflow template that embeds caller-provided paths and settings into shell syntax without shell escaping. The generator does not execute it immediately, but the produced script can run unintended commands if executed with crafted values.
中
scripts/workflow_generator.py:229
Unix shell invocation
script = f"""#!/bin/bash
This starts a generated bash workflow template that embeds caller-provided paths and settings into shell syntax without shell escaping. The generator does not execute it immediately, but the produced script can run unintended commands if executed with crafted values.
中
scripts/workflow_generator.py:278
Unix shell invocation
script = f"""#!/bin/bash
This starts a generated bash workflow template that embeds caller-provided paths and settings into shell syntax without shell escaping. The generator does not execute it immediately, but the produced script can run unintended commands if executed with crafted values.
中
scripts/workflow_generator.py:101
Python file write/append
with open(output_file, 'w') as f:
The generator writes to an output path supplied through the CLI without constraining location or requiring overwrite confirmation. In an agent-driven flow, an unsafe path could overwrite local files.
中
scripts/workflow_generator.py:220
Python file write/append
with open(output_file, 'w') as f:
The generator writes to an output path supplied through the CLI without constraining location or requiring overwrite confirmation. In an agent-driven flow, an unsafe path could overwrite local files.
中
scripts/workflow_generator.py:269
Python file write/append
with open(output_file, 'w') as f:
The generator writes to an output path supplied through the CLI without constraining location or requiring overwrite confirmation. In an agent-driven flow, an unsafe path could overwrite local files.
中
scripts/workflow_generator.py:355
Python file write/append
with open(output_file, 'w') as f:
The generator writes to an output path supplied through the CLI without constraining location or requiring overwrite confirmation. In an agent-driven flow, an unsafe path could overwrite local files.

风险因素

⚙️ 外部命令 (89)
scripts/workflow_generator.py:35 scripts/workflow_generator.py:110 scripts/workflow_generator.py:229 scripts/workflow_generator.py:278 SKILL.md:41-43 SKILL.md:43-51 SKILL.md:51-63 SKILL.md:63-67 SKILL.md:67-71 SKILL.md:71-73 SKILL.md:73-83 SKILL.md:83-96 SKILL.md:96-108 SKILL.md:108-110 SKILL.md:110-116 SKILL.md:116-118 SKILL.md:118-122 SKILL.md:122-124 SKILL.md:124-130 SKILL.md:130-135 SKILL.md:135-137 SKILL.md:137-144 SKILL.md:144-148 SKILL.md:148-151 SKILL.md:151-154 SKILL.md:154-158 SKILL.md:158-163 SKILL.md:163-166 SKILL.md:166-169 SKILL.md:169-173 SKILL.md:173-177 SKILL.md:177-182 SKILL.md:182-190 SKILL.md:190-193 SKILL.md:193-196 SKILL.md:196-200 SKILL.md:200-204 SKILL.md:204-221 SKILL.md:221-229 SKILL.md:229-230 SKILL.md:230-231 SKILL.md:231-232 SKILL.md:232-233 SKILL.md:233-235 SKILL.md:235-242 SKILL.md:242-243 SKILL.md:243 SKILL.md:246-247 SKILL.md:247 SKILL.md:248 SKILL.md:249 SKILL.md:252-253 SKILL.md:253-258 SKILL.md:258-267 SKILL.md:267-274 SKILL.md:274-275 SKILL.md:275-277 SKILL.md:277-282 SKILL.md:282-287 SKILL.md:287-293 SKILL.md:293-304 SKILL.md:304-306 SKILL.md:306-309 SKILL.md:309-310 SKILL.md:310-312 SKILL.md:312-315 SKILL.md:315 SKILL.md:318-323 SKILL.md:323-325 SKILL.md:325-388 SKILL.md:388-391 SKILL.md:391-400 SKILL.md:400-401 SKILL.md:401-402 SKILL.md:402-403 SKILL.md:403-406 SKILL.md:406-418 SKILL.md:418-435 SKILL.md:435-437 SKILL.md:437-443 SKILL.md:443-473 SKILL.md:473-474 SKILL.md:474-475 SKILL.md:475-476 SKILL.md:476-479 SKILL.md:479-488 SKILL.md:488-506 SKILL.md:506-524 SKILL.md:524-525
📁 文件系统访问 (4)
scripts/workflow_generator.py:101 scripts/workflow_generator.py:220 scripts/workflow_generator.py:269 scripts/workflow_generator.py:355
审计者: codex 查看审计历史 →
分享与引用此报告

分享版本化评估报告、中性徽章、嵌入卡片和引用。Skillstore 只报告证据，不判断该 Skill 是否安全。

打开版本化报告
安全评估

复制报告链接

https://skillstore.io/skills/k-dense-ai-deeptools/audits/7?utm_source=security_passport&utm_medium=share&utm_campaign=versioned_report

Markdown 徽章

[![Skillstore security assessment](https://skillstore.io/badges/skills/k-dense-ai-deeptools/security.svg)](https://skillstore.io/skills/k-dense-ai-deeptools?utm_source=security_passport_badge)

HTML 徽章

<a href="https://skillstore.io/skills/k-dense-ai-deeptools?utm_source=security_passport_badge"><img src="https://skillstore.io/badges/skills/k-dense-ai-deeptools/security.svg" alt="Skillstore security assessment" loading="lazy"></a>

嵌入卡片

<iframe src="https://skillstore.io/embed/skills/k-dense-ai-deeptools.html" title="Skillstore Security Assessment" sandbox="allow-popups allow-popups-to-escape-sandbox" loading="lazy" referrerpolicy="no-referrer" width="420" height="180"></iframe>
学术引用 (APA · BibTeX · CFF)

APA 引用

K-Dense-AI. (2026). deeptools security audit report (audit version 7) [Author version unspecified]. Skillstore. https://skillstore.io/skills/k-dense-ai-deeptools/audits/7

BibTeX 引用

@techreport{k-dense-ai-k-dense-ai-deeptools-2026, author = {K-Dense-AI}, title = {deeptools security audit report (audit version 7)}, institution = {Skillstore}, year = {2026}, number = {7}, url = {https://skillstore.io/skills/k-dense-ai-deeptools/audits/7}, note = {Author version unspecified} }

CITATION.cff

cff-version: 1.2.0 message: "If you use this Skill, cite its author and this versioned security audit report." title: "deeptools security audit report (audit version 7)" version: "unspecified" type: report authors: - name: "K-Dense-AI" date-released: "2026-07-05" url: "https://skillstore.io/skills/k-dense-ai-deeptools/audits/7" identifiers: - type: other value: "skillstore:k-dense-ai-deeptools:audit:7" description: "Skillstore immutable audit report identifier"

比较变体

2 个可安装变体

每位作者的技能都保持为独立的可安装项。推荐变体按 Skillstore 证据排序。

为什么此变体排名第一

Skillstore 使用量更高
K-Dense-AI 推荐 当前

k-dense-ai-deeptools

Skillstore 评分 38
证据置信度 中
Skillstore 使用量 15
已更新

2026-09-09

davila7

davila7-deeptools

Skillstore 评分 38
证据置信度 中
Skillstore 使用量 11
已更新

2026-09-09

Skillstore 评分

评分依据 证据置信度: 中
82
架构
85
可维护性
87
内容
70
社区
78
规范符合性

你能构建什么

准备 ChIP-seq 质量控制

创建包含相关性、PCA、覆盖度、片段大小和 fingerprint 输出的质量控制工作流。

构建覆盖度轨道

选择归一化设置，并将已比对的 BAM 文件转换为可直接用于基因组浏览器的 bigWig 轨道。

可视化调控信号

围绕基因、peaks 或其他基因组区域设计 computeMatrix、plotHeatmap 和 plotProfile 步骤。

试试这些提示

启动质量控制工作流 
帮我为一个 input BAM 和两个 ChIP BAM 文件创建一个 deepTools ChIP-seq 质量控制工作流。解释每个输出。
选择归一化方法 
我有来自同一 genome build 的 BAM 文件。请推荐一种用于比较样本间信号的 deepTools 归一化方法。
创建热图计划 
使用我的 bigWig 文件和 gene BED 文件，围绕 TSS 区域设计一个 computeMatrix 和 plotHeatmap 工作流。
审查完整工作流 
审查我计划用于 ATAC-seq 的 deepTools 工作流。检查 Tn5 校正、归一化、片段分析和热图步骤。

最佳实践

  • 在运行 deepTools 命令前，验证 BAM 索引、BED 格式和文件可读性。
  • 根据实验类型和比较目标选择归一化方法。
  • 在处理完整数据集之前，先在较小的基因组区域上测试命令。

避免

  • 不要在未审查文件路径和输出目录的情况下运行生成的 shell 脚本。
  • 不要在 BAM 文件、BED 文件和有效基因组大小值之间混用不同的 genome assemblies。
  • 不要将 ChIP-seq read extension 设置用于 RNA-seq 覆盖度分析。

常见问题

此技能支持哪些实验？
它支持用于 ChIP-seq、RNA-seq、ATAC-seq、MNase-seq 及相关 NGS 信号分析的 deepTools 工作流。
它能生成可运行的脚本吗？
可以。它包含一个工作流生成器，但用户应在执行前审查生成的 shell 脚本。
它会安装 deepTools 吗？
它提供安装指导，但用户环境必须已具备 deepTools 和所需依赖项。
它能验证我的文件吗？
可以。包含的验证脚本会检查文件是否存在、BAM 索引、bigWig 大小和基本 BED 格式。
我应该使用哪种归一化方法？
该技能会根据实验和比较目标解释 RPGC、CPM、RPKM、BPM 和 reads 计数缩放。
它适合自动化执行吗？
由于未强制执行 shell 转义，不应使用不受信任的路径自动运行生成的工作流。

开发者详情

作者

K-Dense-AI

许可证

BSD license

Skillstore 修订

r1

版本提示

作者未声明版本。

仓库

https://github.com/K-Dense-AI/claude-scientific-skills/tree/main/scientific-skills/deeptools

引用

a25199bc7d6b82598536822d1738eb5d5f54025b

维护新鲜度

2026/7/20

使用量

11 次下载 · 192 次浏览

文件结构

📁 assets/

📄 quick_reference.md

📁 references/

📄 effective_genome_sizes.md

📄 normalization_methods.md

📄 tools_reference.md

📄 workflows.md

📁 scripts/

📄 validate_files.py

📄 workflow_generator.py

📄 SKILL.md

K-Dense-AI 的更多技能

查看全部

相关技能

查看全部