菜单
Systematically debug issues using graph-powered code navigation
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。
基于 SOC 职业分类
Work through planned tasks in manageable batches — pick a GitHub issue (plan-labelled, regular, or code-scanning/Dependabot alert), split into parallel subagents, test, review, log, commit, push, open a PR. Never block on CI: start the next batch immediately and reconcile the previous batch's PR (merge if green, fix if red) at the end of the new one. Use when the user asks to work on issues, ship a batch, or process the backlog.
Navigate and understand codebase structure using the knowledge graph
Plan and execute safe refactoring using dependency analysis
Perform a structured code review using change detection and impact
| name | debug-issue |
| description | Systematically debug issues using graph-powered code navigation |
Use the knowledge graph to systematically trace and debug issues.
semantic_search_nodes to find code related to the issue.query_graph with callers_of and callees_of to trace call chains.get_flow to see full execution paths through suspected areas.detect_changes to check if recent changes caused the issue.get_impact_radius on suspected files to see what else is affected.get_minimal_context(task="<your task>") before any other graph tool.detail_level="minimal" on all calls. Only escalate to "standard" when minimal is insufficient.