bug-hunter-stage3-evidence-fusion
bug-hunter 阶段 3 技能。负责对多智能体原始发现做语义去重、桶化聚类与冲突识别,形成可投票的缺陷候选池。
Browse reusable Agent Skills, each with a clear purpose and practical guidance.
bug-hunter 阶段 3 技能。负责对多智能体原始发现做语义去重、桶化聚类与冲突识别,形成可投票的缺陷候选池。
bug-hunter 阶段 4 技能。负责对缺陷桶执行加权共识投票,筛选过阈值问题,并输出裁决级结构化评审报告。
bug-hunter 阶段 2 技能。负责将随机化后的 diff 按 persona 矩阵分发给 8 个子智能体并行评审,并收集统一 JSON 结果。
使用低扰动原子快照、GDB 现场采样和语义对比来调试 DragonOS 内核中的时序问题、Heisenbug、阻塞挂起、丢唤醒和“加日志现象改变”的问题。适用于网络、VFS、调度、IPC、驱动等子系统;当用户提到任务卡住、CPU idle 但请求不返回、阻塞点偶发失效,或明确要求在线取证且不想依赖高频日志时使用。
专用于按照 docs/introduction/develop_nix.md 的流程,通过 Nix dev shell / yolo 命令启动 DragonOS,并在 QEMU nographic 串口中做启动烟雾检查或实时轮询回贴输出。当用户要求“按 develop_nix 跑 yolo”“用 nix yolo 启动 QEMU 看输出”“边跑边轮询输出”“进 guest 后检查 /proc、/sys/fs/cgroup、mount 是否正常”时使用。
通过对比 Linux/gvisor 参考实现来分析 DragonOS gVisor 测试失败。输出结构化的修复文档,包含对于所有失败用例的表格格式分析文档,针对每个具体的失败用例的详细格式修复文档,并提供代码片段。当用户提及 gVisor 测试失败、特定测试用例或询问 bug 分析/修复方案时使用。
Install and validate a .NET SDK from an Azure DevOps internal build. Use this when asked to install, validate, or set up a .NET SDK from a dnceng/internal Azure DevOps build link or build ID.
Author end-to-end Static Web Assets integration tests that exercise MSBuild targets through build, publish, or pack. USE FOR: writing new SWA integration tests, choosing a test asset and base class, dynamically modifying projects at runtime, selecting the right manifest to assert on, verifying the full asset pipeline (primary, compressed, endpoints). DO NOT USE FOR: unit tests of individual tasks (write those directly), baseline regeneration (use swa-baseline-regeneration), troubleshooting failures (use swa-troubleshooting).
Triage a build / compile / restore / WarnAsError failure from its MSBuild binary log. Fetches the binlog (a local build's, or a failed dotnet/fsharp Azure DevOps PR build's published artifact) and analyzes it live via the `binlog-mcp` MCP server — structured errors, root-cause diagnosis, and an MSBuild perf X-ray. NOT for test failures or CheckCodeFormatting: a build binlog has no errors there.
Route code changes to relevant review dimensions based on affected files and folders. Use when reviewing PRs, analyzing code quality, or performing targeted reviews of specific vstest subsystems.