📦
slash-command-encoder
82为 AI 工作流创建斜杠命令
当用户依赖冗长的自然语言提示时，重复性的 AI 工作流会很慢。此技能可帮助定义简洁的斜杠命令，并支持路由、验证、链式调用和帮助文本。
效率 查看
技能 github-multi-repo
📦
github-multi-repo
v1.0.0 内容修订 r1 高风险 ⚙️ 外部命令🌐 网络访问📁 文件系统访问🔑 环境变量
协调多仓库 GitHub 工作流
管理多个仓库的团队需要在相关项目之间保持一致的更新、测试和拉取请求。此技能为协调式多仓库变更提供结构化的 GitHub 工作流指导。
支持: Claude Codex Code(CC)
安装前请停下并请求确认。
请审查安装计划，并在更改文件前获得用户明确同意。
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 "github-multi-repo" from https://skillstore.io/skills/dnyoussef-github-multi-repo.md and its manifest at https://skillstore.io/api/skills/dnyoussef-github-multi-repo/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。
测试它
正在使用“github-multi-repo”。 查找 TypeScript 服务并建议依赖对齐。
预期结果:
- 包含候选服务和共享包的仓库发现摘要。
- 包含测试、分支策略和审查检查点的依赖对齐计划。
- 需要在更新前进行人工检查的仓库风险说明。
正在使用“github-multi-repo”。 在三个仓库之间标准化集成工作流。
预期结果:
- 包含目标文件和预期行为的工作流标准化提案。
- 每个仓库的发布顺序和验证步骤。
- 供审查者和维护者使用的拉取请求检查清单。
正在使用“github-multi-repo”。 准备多仓库安全补丁发布。
预期结果:
- 包含发现、审计、测试和批准阶段的补丁发布计划。
- 用自然语言描述的仓库状态表。
- 更新失败时的回滚和沟通步骤。
安全审计
高风险v8 • 2026/7/5 打开版本化报告
Most static detections are Markdown fences, inline code, or placeholder links and are false positives. Confirmed issues are the examples that direct an agent to run GitHub CLI, use predictable temporary paths, modify hidden configuration, and create pull requests across many repositories. No prompt injection was found, but the community skill needs safety gates for broad repository write operations.
1
已扫描文件
875
分析行数
42
复核项
0
已忽略误报
已确认安全问题 (2)
高
Unbounded Multi-Repository Write Automation
The skill describes cloning many repositories, running package updates, committing, pushing branches, and creating pull requests without an explicit approval gate or allowlist. This can cause broad unintended changes when an agent has organization permissions.
The cited workflows explicitly list repositories, clone them, run package commands, commit changes, push branches, and create pull requests. The risk is the broad write intent, not merely the shell syntax.
中
Absolute Workspace Path Access in Examples
Examples instruct the agent to read and list absolute /workspaces/ruv-FANN paths. In another environment this can read outside the user project or fail unpredictably.
The absolute paths are visible in Read and LS examples and are not parameterized. They appear to be copied from a specific workspace, so legitimate use is possible but portability and boundary risk remain.
能力复核项 (42)
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。
高
Ruby/shell backtick execution
--jq '.[] | select(.name | test("-service$")) | .name' > /tmp/repos.txt`)
This command flow lists repositories, clones them into /tmp, runs npm commands, commits changes, pushes branches, and creates pull requests. That is high-impact automation when the agent has organization permissions.
高
Ruby/shell backtick execution
done`)
This command flow lists repositories, clones them into /tmp, runs npm commands, commits changes, pushes branches, and creates pull requests. That is high-impact automation when the agent has organization permissions.
高
Ruby/shell backtick execution
done`)
This loop installs packages, runs tests, commits, pushes, and opens pull requests across discovered repositories. It is a high-impact write path without an explicit approval gate in the example.
高
Ruby/shell backtick execution
done`)
This patch flow can run npm audit fix, commit changes, push branches, and create pull requests. That is high-impact repository modification under agent credentials.
高
Template literal with command substitution
Bash(`cat /tmp/repos.txt | while read -r repo; do
This command flow lists repositories, clones them into /tmp, runs npm commands, commits changes, pushes branches, and creates pull requests. That is high-impact automation when the agent has organization permissions.
高
Template literal with command substitution
Bash(`for repo in /tmp/audit-*.json; do
This patch flow can run npm audit fix, commit changes, push branches, and create pull requests. That is high-impact repository modification under agent credentials.
高
Hidden file access
path: `${repo}/.github/workflows/integration.yml`,
The example writes a .github workflow file that can change CI behavior across repositories. Hidden workflow changes are high-impact and require review.
中
Ruby/shell backtick execution
--jq '.[] | select(.languages | keys | contains(["TypeScript"]))'`)
This line is part of a Bash-wrapped GitHub CLI discovery flow that enumerates repositories and decodes package metadata. The usage appears legitimate, but it still executes external commands and reads repository content.
中
Ruby/shell backtick execution
done | jq -s '.'`)
This line is part of a Bash-wrapped GitHub CLI discovery flow that enumerates repositories and decodes package metadata. The usage appears legitimate, but it still executes external commands and reads repository content.
中
Ruby/shell backtick execution
-f sha=$(gh api repos/:owner/:repo/git/refs/heads/main --jq '.object.sha')`)
This is part of a GitHub API write workflow that creates refs or updates package content using command substitution. It is intended automation, but it can modify repositories if run with valid credentials.
中
Ruby/shell backtick execution
-f content="$(cat aligned-package.json | base64)"`)
This is part of a GitHub API write workflow that creates refs or updates package content using command substitution. It is intended automation, but it can modify repositories if run with valid credentials.
中
Ruby/shell backtick execution
--jq '.content' | base64 -d > /tmp/claude-source.md`)
This documentation sync example reads repository content into /tmp and sends encoded content back through the GitHub API. The pattern is purposeful, but it performs external commands and repository writes.
中
Ruby/shell backtick execution
-f content="$(cat /tmp/claude-source.md | base64)"`)
This documentation sync example reads repository content into /tmp and sends encoded content back through the GitHub API. The pattern is purposeful, but it performs external commands and repository writes.
中
Ruby/shell backtick execution
- [x] Cross-package compatibility"`)
This line closes a Bash-wrapped gh pr create command that would create a pull request. The exact line is mostly PR body text, but it belongs to an external command block.
中
Ruby/shell backtick execution
--order desc`)
This line is part of a Bash-wrapped gh search command that performs network repository discovery. It is read-only, but it still invokes external tooling outside the local skill.
中
Ruby/shell backtick execution
--json number -q .number`))
This dependency update workflow creates GitHub issues and discovers repositories through nested shell commands. It is legitimate DevOps automation, but it relies on external command execution with organization scope.
中
Ruby/shell backtick execution
done`))
This dependency update workflow creates GitHub issues and discovers repositories through nested shell commands. It is legitimate DevOps automation, but it relies on external command execution with organization scope.
中
Ruby/shell backtick execution
done`)
This security scan example clones repositories and writes audit results into /tmp using shell commands. It is legitimate security work, but it runs external commands across many repositories.
中
Shell command substitution
-f sha=$(gh api repos/:owner/:repo/git/refs/heads/main --jq '.object.sha')`)
This is part of a GitHub API write workflow that creates refs or updates package content using command substitution. It is intended automation, but it can modify repositories if run with valid credentials.
中
Shell command substitution
-f content="$(cat aligned-package.json | base64)"`)
This is part of a GitHub API write workflow that creates refs or updates package content using command substitution. It is intended automation, but it can modify repositories if run with valid credentials.
中
Shell command substitution
-f content="$(cat /tmp/claude-source.md | base64)"`)
This documentation sync example reads repository content into /tmp and sends encoded content back through the GitHub API. The pattern is purposeful, but it performs external commands and repository writes.
中
Shell command substitution
TRACKING_ISSUE=$(Bash(`gh issue create \
This dependency update workflow creates GitHub issues and discovers repositories through nested shell commands. It is legitimate DevOps automation, but it relies on external command execution with organization scope.
中
Shell command substitution
TS_REPOS=$(Bash(`gh repo list org --limit 100 --json name | \
This dependency update workflow creates GitHub issues and discovers repositories through nested shell commands. It is legitimate DevOps automation, but it relies on external command execution with organization scope.
中
Shell command substitution
if [ $(jq '.vulnerabilities | length' $repo) -gt 0 ]; then
The command substitution parses audit file names and vulnerability counts inside an automated patch loop. It is not clearly malicious, but unquoted path-derived values make the shell workflow riskier.
中
Shell command substitution
cd /tmp/$(basename $repo .json | sed 's/audit-//')
The command substitution parses audit file names and vulnerability counts inside an automated patch loop. It is not clearly malicious, but unquoted path-derived values make the shell workflow riskier.
中
Template literal with command substitution
Bash(`gh api repos/:owner/:repo/git/refs \
This is part of a GitHub API write workflow that creates refs or updates package content using command substitution. It is intended automation, but it can modify repositories if run with valid credentials.
中
Template literal with command substitution
Bash(`gh api repos/:owner/:repo/contents/package.json \
This is part of a GitHub API write workflow that creates refs or updates package content using command substitution. It is intended automation, but it can modify repositories if run with valid credentials.
中
Template literal with command substitution
Bash(`gh api repos/:owner/:repo/contents/claude-code-flow/CLAUDE.md \
This documentation sync example reads repository content into /tmp and sends encoded content back through the GitHub API. The pattern is purposeful, but it performs external commands and repository writes.
中
Hidden file access
path: "claude-code-flow/.claude/commands/github/github-modes.md",
The example writes a .claude command file through GitHub tooling. Hidden agent configuration can affect future automation and should be reviewed carefully.
中
Temp directory access
--jq '.[] | select(.name | test("-service$")) | .name' > /tmp/repos.txt`)
The workflow uses predictable /tmp paths while cloning and modifying repositories. This is common in examples, but it can collide with existing files and should use isolated temporary directories.
中
Temp directory access
Bash(`cat /tmp/repos.txt | while read -r repo; do
The workflow uses predictable /tmp paths while cloning and modifying repositories. This is common in examples, but it can collide with existing files and should use isolated temporary directories.
中
Temp directory access
gh repo clone org/$repo /tmp/$repo -- --depth=1
The workflow uses predictable /tmp paths while cloning and modifying repositories. This is common in examples, but it can collide with existing files and should use isolated temporary directories.
中
Temp directory access
cd /tmp/$repo
The workflow uses predictable /tmp paths while cloning and modifying repositories. This is common in examples, but it can collide with existing files and should use isolated temporary directories.
中
Temp directory access
--jq '.content' | base64 -d > /tmp/claude-source.md`)
The documentation sync flow stores repository content in a fixed /tmp file before uploading it. The use is understandable, but fixed temporary paths can expose or overwrite content.
中
Temp directory access
-f content="$(cat /tmp/claude-source.md | base64)"`)
The documentation sync flow stores repository content in a fixed /tmp file before uploading it. The use is understandable, but fixed temporary paths can expose or overwrite content.
中
Temp directory access
gh repo clone org/$repo /tmp/$repo -- --depth=1
The organization-wide update loop clones repositories into predictable /tmp paths and changes directories into them. This creates filesystem and collision risk during broad automation.
中
Temp directory access
cd /tmp/$repo
The organization-wide update loop clones repositories into predictable /tmp paths and changes directories into them. This creates filesystem and collision risk during broad automation.
中
Temp directory access
gh repo clone org/$repo /tmp/$repo -- --depth=1
The security patch workflow clones repositories and writes audit data to predictable /tmp paths before applying fixes. It is legitimate, but broad temp directory use increases operational risk.
中
Temp directory access
cd /tmp/$repo
The security patch workflow clones repositories and writes audit data to predictable /tmp paths before applying fixes. It is legitimate, but broad temp directory use increases operational risk.
中
Temp directory access
npm audit --json > /tmp/audit-$repo.json
The security patch workflow clones repositories and writes audit data to predictable /tmp paths before applying fixes. It is legitimate, but broad temp directory use increases operational risk.
中
Temp directory access
Bash(`for repo in /tmp/audit-*.json; do
The security patch workflow clones repositories and writes audit data to predictable /tmp paths before applying fixes. It is legitimate, but broad temp directory use increases operational risk.
中
Temp directory access
cd /tmp/$(basename $repo .json | sed 's/audit-//')
The security patch workflow clones repositories and writes audit data to predictable /tmp paths before applying fixes. It is legitimate, but broad temp directory use increases operational risk.
风险因素
⚙️ 外部命令 (120)
SKILL.md:42-54 SKILL.md:54-57 SKILL.md:57-63 SKILL.md:63-66 SKILL.md:66-72 SKILL.md:72-79 SKILL.md:79-81 SKILL.md:83-86 SKILL.md:90-98 SKILL.md:98-101 SKILL.md:101-110 SKILL.md:111-114 SKILL.md:130-139 SKILL.md:139-144 SKILL.md:144-160 SKILL.md:162-165 SKILL.md:169-181 SKILL.md:181-184 SKILL.md:184-188 SKILL.md:189-192 SKILL.md:196-204 SKILL.md:204-207 SKILL.md:207-227 SKILL.md:239-240 SKILL.md:240-245 SKILL.md:245-262 SKILL.md:266-278 SKILL.md:278-281 SKILL.md:281-327 SKILL.md:327-330 SKILL.md:330-339 SKILL.md:339-340 SKILL.md:349-354 SKILL.md:354-359 SKILL.md:359-363 SKILL.md:367-370 SKILL.md:376-379 SKILL.md:401-402 SKILL.md:402-405 SKILL.md:405-424 SKILL.md:424-427 SKILL.md:427-431 SKILL.md:436-439 SKILL.md:452-453 SKILL.md:453-458 SKILL.md:458-489 SKILL.md:489-492 SKILL.md:492-509 SKILL.md:509-514 SKILL.md:514-530 SKILL.md:530-533 SKILL.md:533-544 SKILL.md:544-549 SKILL.md:549-560 SKILL.md:560-563 SKILL.md:563-572 SKILL.md:572-575 SKILL.md:575-586 SKILL.md:586-591 SKILL.md:591-597 SKILL.md:597-600 SKILL.md:600-607 SKILL.md:607-610 SKILL.md:610-616 SKILL.md:616-621 SKILL.md:621-648 SKILL.md:648-651 SKILL.md:651-671 SKILL.md:671-676 SKILL.md:676-681 SKILL.md:681-684 SKILL.md:684-689 SKILL.md:689-692 SKILL.md:692-697 SKILL.md:697-732 SKILL.md:732-737 SKILL.md:737-740 SKILL.md:740-745 SKILL.md:745-748 SKILL.md:748-753 SKILL.md:753-758 SKILL.md:758-763 SKILL.md:763-766 SKILL.md:766-771 SKILL.md:771-774 SKILL.md:774-779 SKILL.md:779-784 SKILL.md:784-790 SKILL.md:790-793 SKILL.md:793-799 SKILL.md:799-802 SKILL.md:802-808 SKILL.md:808-813 SKILL.md:813-819 SKILL.md:819-822 SKILL.md:822-828 SKILL.md:828-853 SKILL.md:853-854 SKILL.md:854-855 SKILL.md:855-856 SKILL.md:856-859 SKILL.md:859-860 SKILL.md:860-861 SKILL.md:861-862 SKILL.md:862-868 SKILL.md:124 SKILL.md:162 SKILL.md:169 SKILL.md:196 SKILL.md:363-367 SKILL.md:370-376 SKILL.md:440 SKILL.md:441 SKILL.md:445 SKILL.md:114-130 SKILL.md:160-162 SKILL.md:165-169 SKILL.md:192-196 SKILL.md:359-370 SKILL.md:439-452
🌐 网络访问 (3)
📁 文件系统访问 (17)
🔑 环境变量 (2)
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。
复制报告链接
https://skillstore.io/skills/dnyoussef-github-multi-repo/audits/8?utm_source=security_passport&utm_medium=share&utm_campaign=versioned_reportMarkdown 徽章
[](https://skillstore.io/skills/dnyoussef-github-multi-repo?utm_source=security_passport_badge)HTML 徽章
<a href="https://skillstore.io/skills/dnyoussef-github-multi-repo?utm_source=security_passport_badge"><img src="https://skillstore.io/badges/skills/dnyoussef-github-multi-repo/security.svg" alt="Skillstore security assessment" loading="lazy"></a>嵌入卡片
<iframe src="https://skillstore.io/embed/skills/dnyoussef-github-multi-repo.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
Claude Flow Team. (2026). github-multi-repo security audit report (audit version 8) [Author version 1.0.0]. Skillstore. https://skillstore.io/skills/dnyoussef-github-multi-repo/audits/8BibTeX 引用
@techreport{claude-flow-team-dnyoussef-github-multi-repo-2026,
author = {Claude Flow Team},
title = {github-multi-repo security audit report (audit version 8)},
institution = {Skillstore},
year = {2026},
number = {8},
url = {https://skillstore.io/skills/dnyoussef-github-multi-repo/audits/8},
note = {Author version 1.0.0}
}CITATION.cff
cff-version: 1.2.0
message: "If you use this Skill, cite its author and this versioned security audit report."
title: "github-multi-repo security audit report (audit version 8)"
version: "1.0.0"
type: report
authors:
- name: "Claude Flow Team"
date-released: "2026-07-05"
url: "https://skillstore.io/skills/dnyoussef-github-multi-repo/audits/8"
identifiers:
- type: other
value: "skillstore:dnyoussef-github-multi-repo:audit:8"
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
作者 / 来源 Skillstore 评分 证据置信度 Skillstore 使用量 已更新
dnyoussef-github-multi-repo
Skillstore 评分 38
证据置信度 中
Skillstore 使用量 9
已更新
2026-09-09
Skillstore 评分
评分依据 证据置信度: 中 55
架构
100
可维护性
87
内容
70
社区
74
规范符合性
你能构建什么
协调依赖更新
规划并执行相关服务之间的依赖变更，包括测试和拉取请求。
标准化仓库工作流
在工程组织内对齐模板、工作流文件和文档模式。
监控多仓库健康状况
审查团队之间的仓库连接性、依赖漂移和同步质量。
试试这些提示
发现相关仓库
使用此技能为我的项目识别相关的 GitHub 仓库。在运行命令之前，总结可能的角色、依赖关系和下一步检查。
规划依赖同步
为这些仓库创建依赖同步计划。包括版本检查、测试顺序、分支名称和拉取请求审查步骤。
协调文档更新
比较这些仓库之间的文档模式，并提出一个安全的同步工作流。在我批准之前不要推送更改。
运行受保护的组织更新
设计一个受保护的组织范围更新工作流。使用允许列表、试运行、每个仓库的状态，并在克隆、推送或打开拉取请求前获得明确批准。
最佳实践
- 在进行任何组织范围变更之前，先使用试运行和小范围仓库允许列表。
- 在克隆仓库、安装包、推送分支或创建拉取请求之前，要求人工批准。
- 在命令示例中使用隔离的临时目录、带引号的变量和清理步骤。
避免
- 不要在没有允许列表和审查计划的情况下运行广泛的通配符仓库更新。
- 不要将占位符所有者、路径或 webhook URLs 复制到生产配置中。
- 不要让生成的工作流或隐藏配置更改绕过代码审查。
常见问题
此技能帮助自动化什么？
它帮助规划多个仓库之间的协调式 GitHub 工作流，包括更新、测试、模板和拉取请求。
它需要 GitHub 访问权限吗？
是的。示例假设使用已认证的 GitHub CLI 或具有仓库权限的兼容 GitHub 工具。
它可以在整个组织范围内进行变更吗？
示例展示了组织范围的工作流，但用户应要求允许列表、试运行和明确批准。
直接运行每条命令都安全吗？
不可以。克隆、安装包、推送分支或打开拉取请求的命令在执行前都需要审查。
它提到了哪些工具？
它引用了 claude-flow、ruv-swarm、GitHub CLI、npm、jq 和 GitHub repository APIs。
它可以管理文档同步吗？
可以。它提供了用于同步文档文件并跟踪仓库之间同步状态的示例。
开发者详情
许可证
MIT
作者版本
v1.0.0
Skillstore 修订
r1
引用
c579efd02184f1128d6741d16e6c8d951d7d4208
维护新鲜度
2026/7/18
使用量
5 次下载 · 213 次浏览
文件结构
📄 SKILL.md
Claude Flow Team 的更多技能
📦
micro-skill-creator
82创建聚焦的 AI 微技能
可复用的 AI 工作流常常会变得过于宽泛且难以测试。此技能可帮助定义聚焦的技能，并配备专家提示词、契约和验证计划。
效率 查看
📦
cascade-orchestrator
81编排多模型技能级联
复杂的 AI 工作通常需要多种技能、模型和质量检查。此技能可帮助设计包含路由、sandbox 测试、swarm 协调和记忆的级联。
编码与开发 查看
📦
web-cli-teleport
81选择 Claude Code Web 或 CLI
当团队为任务选择了错误的 Claude Code 界面时，会浪费时间。此技能会推荐 Web、CLI 或混合交接工作流。
效率 查看
📦
agentdb-semantic-vector-search
80构建 AgentDB 语义搜索
团队需要为 RAG 和知识库提供相关的文档检索。此技能指导 AgentDB 设置、嵌入存储、索引、API 搜索和相关性调优。
数据与分析 查看
📦
when-creating-presentations-use-pptx-generation
80创建无障碍 PowerPoint 演示文稿
手动制作演示文稿可能缓慢且不一致。此技能会引导 Claude、Codex 或 Claude Code 完成结构化 PPTX 生成和验证。
办公与报告 查看
相关技能
查看全部📦
incident-command-governor
82构建受治理的事件指挥包
由 yaojingang
事件记录通常缺少清晰的严重性、负责人和面向不同受众的沟通内容。此技能可将运营证据转化为用于响应、复盘和跟进的受治理包。
DevOps 与基础设施 查看
📦
senior-devops
81自动化 DevOps 流水线和部署
由 alirezarezvani
DevOps 工作通常需要可重复的流水线、基础设施和部署模式。此技能提供用于 CI/CD、IaC、部署和运维审查的脚本及参考指南。
DevOps 与基础设施 查看
📦
cloud-devops
80协调云端 DevOps 工作流
由 sickn33
云交付工作通常跨越多种工具和阶段。此技能将基础设施、CI/CD、Kubernetes、监控、安全、成本和恢复工作组织成清晰的提示。
DevOps 与基础设施 查看
📦
terraform-skill
80构建更好的 Terraform 模块
由 sickn33
Terraform 项目常常会逐渐演变成不一致的模块、不清晰的测试和高风险的发布工作流。此技能为 Claude、Codex 和 Claude Code 提供实用的 Terraform 与 OpenTofu 模式，帮助实现更安全的基础设施交付。
DevOps 与基础设施 查看
📦
helm-chart-architect
80设计生产级 Helm Chart 架构
由 92Bilal26
团队经常复制不符合自身运维约束的 Helm 示例。此技能通过依赖项、hooks、values、环境和操作员需求来指导 chart 架构设计。
DevOps 与基础设施 查看
📦
github-cli
79使用 gh 收集 GitHub 上下文
由 BfdCampos
GitHub 工作通常需要 PR、issue、review 和 CI 上下文，而本地 git 无法显示这些内容。此技能指导 Claude、Codex 和 Claude Code 使用 gh 读取命令来获取更丰富的上下文。
DevOps 与基础设施 查看