基于 SOC 职业分类
用 Codex 或 Claude 帮你安装 复制这段 Prompt,粘贴到 Codex、Claude 或其他助手里,让它检查 Skill 页面并帮你完成安装。
直接命令不会经过审查 Prompt;运行前请先检查来源。
npx skills add https://github.com/DataDog/datadog-agent --skill run-jira命令会保持在同一行。复制前请横向滚动并检查完整内容。
想先保存到本地?可下载 SkillsMP 当前能够提供的文件。
正在显示 SKILL.md
| name | run-jira |
| description | Fetch a Jira issue and propose an implementation plan based on codebase analysis |
| argument-hint | <JIRA-KEY e.g. ACTP-1234> |
| model | sonnet |
Fetch the Jira issue $ARGUMENTS from the Datadog Atlassian instance and use it as the basis for a codebase analysis and implementation proposal.
Use the Atlassian MCP tools to fetch the issue. Only request the fields you need to avoid huge responses:
mcp__atlassian__getJiraIssue with:
cloudId: datadoghq.atlassian.netissueIdOrKey: $ARGUMENTSfields: ["summary", "description", "status", "assignee", "issuetype", "comment", "priority"]If the issue cannot be found, stop and inform the user.
Present a clear summary of the Jira issue:
Scan the issue description and comments for links to external resources and fetch them for additional context:
app.datadoghq.com/notebook/<id>): use mcp__datadog-mcp__get_datadog_notebook with the notebook IDgithub.com/.../pull/<number>): use gh pr view <number> via BashWebFetch if accessibleThis step is critical — linked resources often contain the root cause analysis, timelines, and technical details that the Jira description alone does not capture.
Based on the issue requirements and linked resources, explore the codebase to understand:
Use Glob, Grep, and Read tools extensively. For broad exploration, use the Task tool with subagent_type=Explore.
Enter plan mode with EnterPlanMode and write a detailed implementation plan that includes:
Wait for user approval before implementing.
Monitor the current PR's GitLab pipeline to completion, then report success or investigate a failure. Use when the user asks to follow, babysit, watch, or wait on a PR/pipeline, or just after pushing to / creating a PR.
Classify a failed CI as either caused by an active incident, flakiness, or a true code regression. Use when a PR's pipeline is red and it isn't obvious whether the PR's own changes are at fault. Trigger phrases include: - "investigate this CI failure" - "please fix CI" - "why did this job fail" - "is there an incident affecting CI" - "should I retry this" This should also be invoked whenever the user asks you to investigate _or fix_ a failing CI, to ensure we don't spend hours trying to fix something broken upstream.
Write or extend Datadog Agent new-e2e tests, including fakeintake coverage and the GitLab wiring that runs them; derives scope from the current diff when no target is named. Not for running tests that already exist (run-e2e, run-windows-e2e), or for judging whether a behavior belongs in E2E at all (e2e-audit).