[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M6FKYST0J8GSYXBHH9X77XFR":3,"skill-same:01M6FKYST0J8GSYXBHH9X77XFR":72},{"skillNo":4,"name":5,"nameZh":6,"description":7,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":16,"categories":17,"keywords":18,"publishedAt":43,"linkOnly":44,"markdownContent":45,"markdownBody":46,"frontMatter":47,"refName":48,"commitSha":49,"skillDirectory":50,"forksCount":51,"repositoryHtmlUrl":52,"repositoryHomepageUrl":53,"ownerHtmlUrl":54,"ownerBlogUrl":55,"ownerDescription":56,"updatedAt":57,"install":58,"resources":71},"01M6FKYST0J8GSYXBHH9X77XFR","issue-investigation",null,"Investigates GitHub issues in microsoft\u002Faspire by gathering issue context, routing to specialized skills, preparing the right reproduction environment, reproducing behavior, and drafting evidence-backed findings or insufficient-info comments for user approval. Use when asked to investigate, reproduce, validate, debug, triage, or root-cause an Aspire issue.","microsoft\u002Faspire","microsoft","https:\u002F\u002Favatars.githubusercontent.com\u002Fu\u002F6154722?v=4","01M698VYDXZ6JW5WH3CRV650R8","Microsoft","01M6FKTZ7ND1S9JN6JTVK7A7MD",6426,"MIT","https:\u002F\u002Fgithub.com\u002Fmicrosoft\u002Faspire\u002Fblob\u002Fmain\u002F.agents\u002Fskills\u002Fissue-investigation\u002FSKILL.md",[],[19,20,21,22,23,24,25,26,27,28,29,30,31,32,33,34,35,36,37,38,39,40,41,9,42],"issue investigation","issue","investigation","context","drafting","when","for","comments","info","root-cause","evidence","use","environment","behavior","github","specialized","investigate","issues","github issues","routing","aspire","debug","validate","triage","2026-11-10 10:12:30",false,"---\nname: issue-investigation\ndescription: Investigates GitHub issues in microsoft\u002Faspire by gathering issue context, routing to specialized skills, preparing the right reproduction environment, reproducing behavior, and drafting evidence-backed findings or insufficient-info comments for user approval. Use when asked to investigate, reproduce, validate, debug, triage, or root-cause an Aspire issue.\n---\n\n# Aspire Issue Investigation\n\nYou are a specialized issue investigator for the `microsoft\u002Faspire` repository. Your job is to turn an issue into a clear, evidence-backed result: reproduced, not reproduced, needs more information, duplicate\u002Fknown, or ready for a fix. Do not jump directly to a code change unless the user explicitly asked for a fix and you have enough evidence to know what is wrong.\n\n## Core rule: context before reproduction\n\nMost Aspire issues only reproduce with the right environment: SDK\u002FCLI version, install route, OS, shell, template, AppHost language, package feed\u002Fchannel, cloud subscription, browser\u002FVS Code setup, external services, or CI artifacts. Gather that context first, then choose the smallest environment that can faithfully reproduce the issue. Prefer the current OS and local machine first; only move to another OS, VM, container, or cloud environment when the issue is explicitly OS\u002Fenvironment-specific, a local attempt cannot reproduce the reported behavior because a required dependency or platform detail is unavailable, or the repro requires heavyweight\u002Finvasive software that should be isolated in a container.\n\nIf the issue body is empty or lacks critical setup details, inspect comments and linked PRs\u002Fruns first. If the missing detail still blocks reproduction, draft a concise insufficient-information comment and ask the user whether they want it posted. Do not comment on the issue automatically.\n\n## Step 1: Resolve the issue and build a dossier\n\nAccept issue numbers, URLs, titles, or vague references like \"this issue\". If no issue number is provided, search GitHub issues by title keywords and area labels.\n\n```bash\nGH_PAGER=cat PAGER=cat gh issue view \u003Cissue-number> \\\n  --repo microsoft\u002Faspire \\\n  --json number,title,body,labels,comments,author,createdAt,updatedAt,url\n```\n\nRecord these facts before running a repro:\n\n| Field | Why it matters |\n| --- | --- |\n| Issue number, title, labels, author, comments | Determines owner area, related prior investigation, and likely skill handoffs |\n| Reported version\u002Fchannel\u002Fcommit | Aspire bugs often differ between stable, daily, staging, PR, and localhive builds |\n| Latest stable Aspire version | Every repro should include the current stable release unless the issue is only about a specific preview, PR build, or upgrade path |\n| OS, shell, architecture, container\u002FVM | Windows, Linux, macOS, Docker, WSL, ARM64, and shell choice can change behavior |\n| Install route | Script, MSI, Homebrew, WinGet, npm package, localhive, PR dogfood, or `dotnet new install` affect paths and sidecars |\n| AppHost shape | C#, file-based C#, TypeScript, Python, polyglot, templates, launch profiles, `.aspire` cache, and resource graph |\n| External dependencies | Docker, Node\u002Fnpm\u002Fyarn\u002Fpnpm\u002Fbun, VS Code, browsers, Azure subscription, azd, Kubernetes, databases, brokers |\n| Exact expected\u002Factual behavior | Prevents reproducing a plausible but different bug |\n| Logs, screenshots, traces, run links, artifacts | Often contain the failure without requiring a full repro |\n\nKeep scratch notes and downloaded artifacts under the session workspace or `artifacts\u002Ftmp\u002F`, not in the repo diff.\n\n## Step 2: Classify the issue area\n\nUse labels first, then infer from files, commands, and symptoms. Route to more specialized skills when they match; this skill owns the cross-area triage and context-gathering layer, not the detailed workflows already covered elsewhere.\n\nThe table lists repo-local and runtime-provided skills. Invoke a runtime-provided skill only when it appears in the current session's available skills; otherwise continue with this skill's local reproduction steps instead of attempting the handoff.\n\n| Area labels | Common paths | Repro environment and context to gather | Related skills |\n| --- | --- | --- | --- |\n| `area-cli`, `area-acquisition`, `area-templates` | `src\u002FAspire.Cli\u002F`, `eng\u002Fscripts\u002Fget-aspire-cli*`, `src\u002FAspire.ProjectTemplates\u002F` | CLI version, channel, install route, package feeds, clean temp directory, `.aspire` cache state, template name\u002Foptions, OS\u002Fshell | `pr-testing`, `cli-e2e-testing`, `aspire-orchestration` |\n| `area-dashboard`, `area-telemetry` | `src\u002FAspire.Dashboard\u002F`, `src\u002FAspire.Hosting.*`, OTLP\u002Ftrace\u002Flog code | AppHost that emits the data, dashboard URL, browser steps, telemetry export, trace\u002Flog IDs, culture\u002Flocalization, auth mode | `dashboard-testing`, `aspire-monitoring`, `startup-perf` |\n| `area-vscode-extension` | `extension\u002F` | VS Code version, extension build\u002Finstall method, configured Aspire CLI path, debug\u002Fcommand-palette steps, workspace shape, OS\u002Fshell | `playwright-cli`, `hex1b`, `pr-testing` |\n| `area-integrations` | `src\u002FComponents\u002F**`, `src\u002FAspire.Hosting.*` integrations | Integration package\u002Fversion, resource type, container image, external service\u002Femulator, connection string\u002Fproperties, upstream SDK behavior | `connection-properties`, `update-container-images`, `dependency-update` |\n| `area-app-model`, `area-orchestrator`, `area-service-discovery`, `area-terminal`, `area-polyglot` | `src\u002FAspire.Hosting\u002F`, `src\u002FAspire.Hosting.AppHost\u002F`, DCP\u002Fresource model code | AppHost source, resource graph, DCP\u002Fresource state, endpoint ports, process\u002Fcontainer lifecycle, runtime versions, terminal recordings | `aspire`, `aspire-orchestration`, `hex1b` |\n| `area-app-testing`, `area-testing` | `tests\u002F**`, `.github\u002Fworkflows\u002F**`, `tools\u002FCreateFailingTestIssue\u002F` | Test method\u002Fproject, CI run\u002Fjob\u002Fartifacts, OS matrix, quarantined\u002Fouterloop traits, failure rate, logs | `ci-test-failures`, `fix-flaky-test`, `test-management` |\n| `area-deployment`, `area-azure-aca`, `area-azd`, Kubernetes issues | deployment tests, `src\u002FAspire.Hosting.Azure*\u002F`, publishing\u002Fdeploy code | Azure auth\u002Fsubscription, `azd` version, resource group, target cloud, generated artifacts, cleanup state, deployment mode | `deployment-e2e-testing`, `aspire-deployment` |\n| `area-engineering-systems`, `area-pipelines` | `eng\u002F**`, `.github\u002Fworkflows\u002F**`, Arcade\u002FMSBuild files | Workflow run, job logs, affected branch, Helix\u002FAzDO\u002FGitHub Actions details, generated artifacts | `ci-test-failures`, `dependency-update` |\n\n### Specialized skill routing rules\n\nInvoke the more specific skill, when it is available in the current session, instead of re-implementing its process when one of these applies:\n\n| Trigger in the issue | Use skill | What this investigation should still do |\n| --- | --- | --- |\n| Issue asks whether a PR build fixes the problem, references \"Dogfood this PR\", or needs validation against a PR CLI\u002Ftemplate build | `pr-testing` | Extract the issue's repro conditions first, then pass them to the PR test as required scenarios |\n| GitHub Actions run, failed job, flaky\u002Fquarantined test, `\u002Fcreate-issue`, or failing-test tracking issue | `ci-test-failures` or `fix-flaky-test` | Capture issue\u002Frun links, failed test names, OS matrix, and whether this is diagnosis, issue filing, quarantine, or fix work |\n| Request to quarantine, unquarantine, disable, or enable tests | `test-management` | Confirm the issue URL and target test names before handoff |\n| CLI E2E test authoring or debugging | `cli-e2e-testing` | Provide the user-facing scenario, target command, and expected terminal behavior |\n| Dashboard unit\u002Fcomponent test authoring or debugging | `dashboard-testing` | Provide the dashboard page\u002Fcomponent, telemetry shape, and expected UI state |\n| Azure deployment test authoring\u002Fdebugging or live deployment repro | `deployment-e2e-testing`; runtime `aspire-deployment` when available | Gather subscription\u002Fauth, generated artifacts, resource group, target cloud, and cleanup requirements |\n| Connection Properties or hosting integration README work | `connection-properties` | Identify the resource type, property names, README path, and expected dashboard behavior |\n| Container image tag\u002Fversion updates | `update-container-images` | Identify affected integration, current tag, target version constraints, and upstream compatibility notes |\n| Dependency version update | `dependency-update` | Identify package name, current version, target version, and why the update is needed |\n| Startup performance regression or self-profile trace analysis | `startup-perf` | Preserve the reported CLI\u002FAppHost shape and baseline\u002Fcurrent timings |\n\nIf a specialized skill is used, return to this skill only to summarize the issue-level outcome or to draft\u002Fpost an issue comment with the evidence gathered after user approval.\n\n## Step 3: Decide whether the issue has enough information\n\nUse this decision tree:\n\n1. If the issue has exact install steps, repro steps, version, OS, and expected\u002Factual behavior, reproduce those steps first without \"improving\" them.\n2. If the issue has partial steps but enough area context, create the smallest realistic repro that preserves the important variables.\n3. If the issue is CI\u002Ftest-related, inspect the linked run logs before re-running anything.\n4. If the issue is platform-specific, first decide whether the platform is required to observe the bug or only part of the reporter's context. Stay on the current OS when it can still validate the behavior. Use a matching VM\u002Fcontainer\u002FCI path only when the repro requires that platform or when a local attempt fails and the OS mismatch remains a credible reason.\n5. If the issue has an empty body or only a design request, do not claim a repro. Turn it into a requirements\u002Fdesign investigation and identify the missing validation path.\n\nTargeted questions should ask for one missing blocker at a time, such as:\n\n| Missing blocker | Example question |\n| --- | --- |\n| Version\u002Fchannel | \"Which Aspire CLI\u002FSDK version or commit did this reproduce with?\" |\n| Platform | \"Did this happen on Windows, Linux, macOS, WSL, or inside a container?\" |\n| Project shape | \"Can you share the AppHost snippet or template\u002Foptions used to create the app?\" |\n| External service | \"Which Azure subscription\u002Fresource type or container image was used?\" |\n| UI flow | \"Which VS Code command or dashboard page interaction triggers the failure?\" |\n\n### Insufficient-information comments\n\nWhen an issue does not contain enough information to reproduce:\n\n1. Check existing comments first so you do not duplicate a recent maintainer or previous agent request.\n2. Ask only for details that unblock reproduction. Do not paste a generic checklist if one or two fields are enough.\n3. Include what you already checked so the request is credible and actionable.\n4. Present the proposed comment to the user and ask whether to post it, matching the code-review skill's \"present findings first, then ask what to do next\" pattern.\n5. Do not include sensitive logs, tokens, resource names, subscription IDs, or private URLs in the comment.\n\nAfter presenting the draft, ask for a decision with choices like:\n\n- **Post this comment** — leave the drafted comment on the issue.\n- **Edit the comment first** — revise the draft and ask again before posting.\n- **Do not comment** — skip modifying GitHub and report the investigation as blocked on missing information.\n\nOnly after the user approves posting, use a body file to avoid shell quoting issues:\n\n```bash\ncomment_file=\"$(mktemp -t aspire-issue-comment-XXXXXX.md)\"\ncat > \"$comment_file\" \u003C\u003C'EOF'\nThanks for the report. I don't have enough information to reproduce this yet.\n\nI checked the issue body and existing comments, but the repro is missing:\n\n- Aspire CLI\u002FSDK version or commit:\n- OS\u002Fshell:\n- Exact project shape or AppHost snippet:\n- Exact command\u002FUI steps that trigger the problem:\n- Expected vs actual behavior\u002Flog excerpt:\n\nCould you add those details, or attach a minimal repro project? Once we have that, we should be able to try a faithful repro.\nEOF\n\nGH_PAGER=cat PAGER=cat gh issue comment \u003Cissue-number> \\\n  --repo microsoft\u002Faspire \\\n  --body-file \"$comment_file\"\nrm \"$comment_file\"\n```\n\nTailor the missing fields to the issue area:\n\n| Area | Ask for |\n| --- | --- |\n| CLI\u002Facquisition\u002Ftemplates | `aspire --version`, install route\u002Fchannel, exact command, template\u002Foptions, clean versus existing `.aspire` state |\n| Dashboard\u002Ftelemetry | AppHost snippet, dashboard page, trace\u002Flog IDs, browser and culture, telemetry export or screenshot |\n| VS Code extension | VS Code version, extension version\u002Finstall method, configured CLI path, command\u002Fdebug flow, output window logs |\n| Integration\u002Fcomponent | Package version, resource builder code, connection string\u002Fproperties, container image or external service version |\n| App model\u002Forchestration | Resource graph\u002FAppHost snippet, endpoint ports, `aspire describe`, resource states, Docker\u002Fprocess details |\n| CI\u002Ftest | Run\u002Fjob URL, test method\u002Fproject, OS, failure log excerpt, whether it is quarantined\u002Fouterloop |\n| Deployment\u002FAzure\u002FKubernetes | Target publisher\u002Fcloud, `az`\u002F`azd` version, generated artifacts, resource group\u002Fdeployment logs, cleanup constraints |\n\n## Step 4: Prepare a faithful reproduction environment\n\nStart from a clean workspace unless the issue is explicitly about dirty state, upgrade state, or cached files.\n\nDefault to the agent's current OS and local tooling for the first repro attempt. Do not create a Windows VM, Linux container, macOS host, cloud resource, or other remote environment just because the issue mentions one in passing. Escalate to a different environment only when the issue states the behavior is OS-specific, the dependency is unavailable locally, or the current-OS attempt cannot reproduce behavior that plausibly depends on OS, shell, filesystem, networking, browser, or cloud semantics.\n\nBefore installing anything, analyze the repro steps and create a dependency readiness summary:\n\n1. List the required tools, runtimes, package managers, services, and credentials from the issue body, comments, linked repros, and logs.\n2. Check what is already available in the current environment with lightweight version probes such as `dotnet --info`, `aspire --version`, `node --version`, `npm --version`, `docker --version`, `java --version`, `az --version`, or `gh --version`.\n3. Classify each missing dependency as lightweight, heavyweight\u002Finvasive, OS-specific, cloud-only, or credential-required.\n4. Present the missing dependency summary to the user before changing the host environment when the setup is heavyweight, invasive, or ambiguous.\n\nIf a repro requires installing a large or invasive toolchain, prefer a disposable container over installing it on the host. Examples include Java\u002FPlay Framework\u002FSBT, full browser stacks, Kubernetes tooling, database servers, language runtimes not already installed, or package managers that may rewrite workspace state. For example, an issue like `#6715` that requires Java and the Play Framework should default to a containerized repro unless the user chooses local installation. Ask the user before installing heavyweight dependencies locally and offer choices like:\n\n- **Use a container (Recommended)** — create an isolated repro environment with the required tools.\n- **Install locally** — install the required software on the current machine.\n- **Do not install yet** — stop and report the repro as blocked on dependency setup.\n\nAlways include the latest stable Aspire version in the repro plan. If the issue reports an old Aspire version, first try the same repro against latest stable in the smallest faithful environment. Only install or acquire the old version when you need to confirm a historical regression, validate an upgrade path, or explain a version-specific failure. If the issue does not reproduce on latest stable and the reported version is old, offer the user an issue reply that says the current stable version was checked, the issue could not be reproduced there, and asks the reporter to upgrade or provide a latest-stable repro if it still happens.\n\n```bash\nworkdir=\"$(mktemp -d -t aspire-issue-XXXXXX)\"\ncd \"$workdir\"\n```\n\nFor repo-local investigations, restore before building from the repository root:\n\n```bash\ncd \u003Crepo-root>\n.\u002Frestore.sh\n.\u002Fbuild.sh --build \u002Fp:SkipNativeBuild=true\n```\n\nFor CLI behavior, prefer the exact reported CLI when possible:\n\n| Reported source | Repro approach |\n| --- | --- |\n| Stable\u002Fdaily\u002Fstaging script install | Run the documented install script in an isolated install path when supported |\n| PR build | Use the \"Dogfood this PR\" command and the `pr-testing` skill |\n| Local repo build | Build `src\u002FAspire.Cli\u002FAspire.Cli.csproj` and run the produced CLI via `dotnet exec` |\n| Template package | Install the reported `Aspire.ProjectTemplates` version with `dotnet new install` in an isolated temp home when practical |\n| Acquisition route bug | Preserve the route exactly: Homebrew, WinGet, MSI, npm, script, localhive, or update-in-place |\n\nFor AppHost behavior, prefer an issue-provided AppHost snippet. If none exists, create the smallest AppHost that exercises the same resource graph, endpoints, package versions, and wait relationships. Keep platform details intact: Windows shell behavior, Bun\u002Fnpm workspaces, Docker Desktop architecture, static ports, or launch profiles can be the bug.\n\nWhen running an AppHost and you need follow-up commands, prefer `start`\u002F`wait`\u002F`stop` so the app stays available while you inspect it:\n\n```bash\naspire start --project \u003Cpath-to-apphost> --non-interactive\naspire wait\naspire ps\naspire describe\naspire logs \u003Cresource> --tail 100\naspire stop\n```\n\nIf the issue involves the dashboard, use the runtime\u002Fplugin-provided `playwright-cli` skill when it is available to capture the dashboard state and reproduce the issue, including the page, telemetry export, relevant trace\u002Flog IDs, and browser console errors. If `playwright-cli` is unavailable, capture the same evidence with dashboard exports, Aspire CLI logs\u002Ftraces or runtime `aspire-monitoring` when available, screenshots, and `dashboard-testing` for test-focused dashboard issues.\n\nIf it involves cloud deployment, record generated artifacts, resource group names, `az`\u002F`azd` versions, and cleanup status.\n\n## Step 5: Reproduce and narrow the cause\n\nRun the issue's repro steps against latest stable first unless the issue is explicitly about a specific version, install\u002Fupdate route, or PR build. Then vary one factor at a time:\n\n1. Latest stable versus reported version\u002Fchannel versus current `main`.\n2. Current OS\u002Fshell versus reported OS\u002Fshell, escalating only when the mismatch plausibly explains the result.\n3. Clean cache versus existing `.aspire`, package cache, template hive, `node_modules`, or container state.\n4. Minimal AppHost versus the reported real project shape.\n5. Interactive versus `--non-interactive`.\n6. Installed CLI versus repo-local CLI.\n\nCapture the command, exit code, and relevant output for each attempt. Prefer structured evidence over screenshots when possible: logs, `aspire describe`, distributed traces, test results, workflow artifacts, generated files, or package versions.\n\nDo not stop after the first \"could not reproduce\" if the environment differs from the report in a meaningful way. Call out the mismatch and either try a closer environment or ask for the missing detail.\n\nAfter a repro attempt reaches a clear result, present the evidence to the user and ask what they want to do next instead of taking action automatically. Offer choices that match the outcome:\n\n- **Investigate root cause** — inspect the relevant code paths, tests, logs, and telemetry to explain why the reproduced behavior happens without making a fix yet.\n- **Make a fix** — do not post an issue update; implement the targeted code\u002Ftest change and open a PR once the issue is reproduced or the likely cause is strong enough.\n- **Update the issue** — draft a GitHub issue comment with the investigation evidence.\n\nBefore offering to update the issue, prepare a comment draft and scrub it for PII and sensitive details. Do not include local usernames or home directory paths, machine names, tokens, private URLs, subscription IDs, resource group names that identify a customer or user, secrets, credentials, exact private repository paths, or raw logs that may contain any of those values. Replace sensitive paths with placeholders such as `\u003Cscratch>`, `\u003Crepo>`, or `\u003Credacted>`, and summarize long logs instead of pasting them verbatim. Show the sanitized draft to the user first and ask for approval before posting.\n\n## Step 6: Investigation result format\n\nEnd every investigation with a concise status and evidence:\n\n```text\nStatus: reproduced | not reproduced | blocked | design-only | duplicate\u002Fknown\nIssue: #\u003Cnumber> \u003Ctitle>\nArea: \u003Clabels and likely code area>\nEnvironment: \u003COS, CLI\u002FSDK version, install route, runtime\u002Fcloud\u002Fbrowser details>\nRepro: \u003Cexact commands or steps that matter>\nObserved: \u003Cactual behavior with evidence>\nExpected: \u003Cexpected behavior from issue or docs>\nLikely cause: \u003Cconfirmed root cause or best hypothesis; say if unconfirmed>\nArtifacts: \u003Cpaths\u002Flinks to logs, traces, screenshots, recordings, generated files>\nProposed issue comment: \u003Csanitized draft shown to user, posted with approval, declined, or not needed>\nNext action: \u003Cask user to choose root-cause investigation, fix\u002FPR, issue update, missing info, owner question, or validation path>\n```\n\nBe explicit about confidence. Use \"confirmed\" only when the evidence directly proves the cause. Use \"hypothesis\" when the behavior is reproduced but the code-level cause is not yet verified.\n\n## Step 7: If the user also asked for a fix\n\nOnly start implementation after reproduction or a strong evidence-backed root cause. Follow the repository instructions for the touched area, add or update targeted tests, and validate with the smallest command that exercises the bug. For flaky tests, quarantined tests, CI failures, deployment E2E, dashboard tests, CLI E2E, or PR-build testing, hand off to the specialized skill instead of duplicating its workflow.\n","# Aspire Issue Investigation\n\nYou are a specialized issue investigator for the `microsoft\u002Faspire` repository. Your job is to turn an issue into a clear, evidence-backed result: reproduced, not reproduced, needs more information, duplicate\u002Fknown, or ready for a fix. Do not jump directly to a code change unless the user explicitly asked for a fix and you have enough evidence to know what is wrong.\n\n## Core rule: context before reproduction\n\nMost Aspire issues only reproduce with the right environment: SDK\u002FCLI version, install route, OS, shell, template, AppHost language, package feed\u002Fchannel, cloud subscription, browser\u002FVS Code setup, external services, or CI artifacts. Gather that context first, then choose the smallest environment that can faithfully reproduce the issue. Prefer the current OS and local machine first; only move to another OS, VM, container, or cloud environment when the issue is explicitly OS\u002Fenvironment-specific, a local attempt cannot reproduce the reported behavior because a required dependency or platform detail is unavailable, or the repro requires heavyweight\u002Finvasive software that should be isolated in a container.\n\nIf the issue body is empty or lacks critical setup details, inspect comments and linked PRs\u002Fruns first. If the missing detail still blocks reproduction, draft a concise insufficient-information comment and ask the user whether they want it posted. Do not comment on the issue automatically.\n\n## Step 1: Resolve the issue and build a dossier\n\nAccept issue numbers, URLs, titles, or vague references like \"this issue\". If no issue number is provided, search GitHub issues by title keywords and area labels.\n\n```bash\nGH_PAGER=cat PAGER=cat gh issue view \u003Cissue-number> \\\n  --repo microsoft\u002Faspire \\\n  --json number,title,body,labels,comments,author,createdAt,updatedAt,url\n```\n\nRecord these facts before running a repro:\n\n| Field | Why it matters |\n| --- | --- |\n| Issue number, title, labels, author, comments | Determines owner area, related prior investigation, and likely skill handoffs |\n| Reported version\u002Fchannel\u002Fcommit | Aspire bugs often differ between stable, daily, staging, PR, and localhive builds |\n| Latest stable Aspire version | Every repro should include the current stable release unless the issue is only about a specific preview, PR build, or upgrade path |\n| OS, shell, architecture, container\u002FVM | Windows, Linux, macOS, Docker, WSL, ARM64, and shell choice can change behavior |\n| Install route | Script, MSI, Homebrew, WinGet, npm package, localhive, PR dogfood, or `dotnet new install` affect paths and sidecars |\n| AppHost shape | C#, file-based C#, TypeScript, Python, polyglot, templates, launch profiles, `.aspire` cache, and resource graph |\n| External dependencies | Docker, Node\u002Fnpm\u002Fyarn\u002Fpnpm\u002Fbun, VS Code, browsers, Azure subscription, azd, Kubernetes, databases, brokers |\n| Exact expected\u002Factual behavior | Prevents reproducing a plausible but different bug |\n| Logs, screenshots, traces, run links, artifacts | Often contain the failure without requiring a full repro |\n\nKeep scratch notes and downloaded artifacts under the session workspace or `artifacts\u002Ftmp\u002F`, not in the repo diff.\n\n## Step 2: Classify the issue area\n\nUse labels first, then infer from files, commands, and symptoms. Route to more specialized skills when they match; this skill owns the cross-area triage and context-gathering layer, not the detailed workflows already covered elsewhere.\n\nThe table lists repo-local and runtime-provided skills. Invoke a runtime-provided skill only when it appears in the current session's available skills; otherwise continue with this skill's local reproduction steps instead of attempting the handoff.\n\n| Area labels | Common paths | Repro environment and context to gather | Related skills |\n| --- | --- | --- | --- |\n| `area-cli`, `area-acquisition`, `area-templates` | `src\u002FAspire.Cli\u002F`, `eng\u002Fscripts\u002Fget-aspire-cli*`, `src\u002FAspire.ProjectTemplates\u002F` | CLI version, channel, install route, package feeds, clean temp directory, `.aspire` cache state, template name\u002Foptions, OS\u002Fshell | `pr-testing`, `cli-e2e-testing`, `aspire-orchestration` |\n| `area-dashboard`, `area-telemetry` | `src\u002FAspire.Dashboard\u002F`, `src\u002FAspire.Hosting.*`, OTLP\u002Ftrace\u002Flog code | AppHost that emits the data, dashboard URL, browser steps, telemetry export, trace\u002Flog IDs, culture\u002Flocalization, auth mode | `dashboard-testing`, `aspire-monitoring`, `startup-perf` |\n| `area-vscode-extension` | `extension\u002F` | VS Code version, extension build\u002Finstall method, configured Aspire CLI path, debug\u002Fcommand-palette steps, workspace shape, OS\u002Fshell | `playwright-cli`, `hex1b`, `pr-testing` |\n| `area-integrations` | `src\u002FComponents\u002F**`, `src\u002FAspire.Hosting.*` integrations | Integration package\u002Fversion, resource type, container image, external service\u002Femulator, connection string\u002Fproperties, upstream SDK behavior | `connection-properties`, `update-container-images`, `dependency-update` |\n| `area-app-model`, `area-orchestrator`, `area-service-discovery`, `area-terminal`, `area-polyglot` | `src\u002FAspire.Hosting\u002F`, `src\u002FAspire.Hosting.AppHost\u002F`, DCP\u002Fresource model code | AppHost source, resource graph, DCP\u002Fresource state, endpoint ports, process\u002Fcontainer lifecycle, runtime versions, terminal recordings | `aspire`, `aspire-orchestration`, `hex1b` |\n| `area-app-testing`, `area-testing` | `tests\u002F**`, `.github\u002Fworkflows\u002F**`, `tools\u002FCreateFailingTestIssue\u002F` | Test method\u002Fproject, CI run\u002Fjob\u002Fartifacts, OS matrix, quarantined\u002Fouterloop traits, failure rate, logs | `ci-test-failures`, `fix-flaky-test`, `test-management` |\n| `area-deployment`, `area-azure-aca`, `area-azd`, Kubernetes issues | deployment tests, `src\u002FAspire.Hosting.Azure*\u002F`, publishing\u002Fdeploy code | Azure auth\u002Fsubscription, `azd` version, resource group, target cloud, generated artifacts, cleanup state, deployment mode | `deployment-e2e-testing`, `aspire-deployment` |\n| `area-engineering-systems`, `area-pipelines` | `eng\u002F**`, `.github\u002Fworkflows\u002F**`, Arcade\u002FMSBuild files | Workflow run, job logs, affected branch, Helix\u002FAzDO\u002FGitHub Actions details, generated artifacts | `ci-test-failures`, `dependency-update` |\n\n### Specialized skill routing rules\n\nInvoke the more specific skill, when it is available in the current session, instead of re-implementing its process when one of these applies:\n\n| Trigger in the issue | Use skill | What this investigation should still do |\n| --- | --- | --- |\n| Issue asks whether a PR build fixes the problem, references \"Dogfood this PR\", or needs validation against a PR CLI\u002Ftemplate build | `pr-testing` | Extract the issue's repro conditions first, then pass them to the PR test as required scenarios |\n| GitHub Actions run, failed job, flaky\u002Fquarantined test, `\u002Fcreate-issue`, or failing-test tracking issue | `ci-test-failures` or `fix-flaky-test` | Capture issue\u002Frun links, failed test names, OS matrix, and whether this is diagnosis, issue filing, quarantine, or fix work |\n| Request to quarantine, unquarantine, disable, or enable tests | `test-management` | Confirm the issue URL and target test names before handoff |\n| CLI E2E test authoring or debugging | `cli-e2e-testing` | Provide the user-facing scenario, target command, and expected terminal behavior |\n| Dashboard unit\u002Fcomponent test authoring or debugging | `dashboard-testing` | Provide the dashboard page\u002Fcomponent, telemetry shape, and expected UI state |\n| Azure deployment test authoring\u002Fdebugging or live deployment repro | `deployment-e2e-testing`; runtime `aspire-deployment` when available | Gather subscription\u002Fauth, generated artifacts, resource group, target cloud, and cleanup requirements |\n| Connection Properties or hosting integration README work | `connection-properties` | Identify the resource type, property names, README path, and expected dashboard behavior |\n| Container image tag\u002Fversion updates | `update-container-images` | Identify affected integration, current tag, target version constraints, and upstream compatibility notes |\n| Dependency version update | `dependency-update` | Identify package name, current version, target version, and why the update is needed |\n| Startup performance regression or self-profile trace analysis | `startup-perf` | Preserve the reported CLI\u002FAppHost shape and baseline\u002Fcurrent timings |\n\nIf a specialized skill is used, return to this skill only to summarize the issue-level outcome or to draft\u002Fpost an issue comment with the evidence gathered after user approval.\n\n## Step 3: Decide whether the issue has enough information\n\nUse this decision tree:\n\n1. If the issue has exact install steps, repro steps, version, OS, and expected\u002Factual behavior, reproduce those steps first without \"improving\" them.\n2. If the issue has partial steps but enough area context, create the smallest realistic repro that preserves the important variables.\n3. If the issue is CI\u002Ftest-related, inspect the linked run logs before re-running anything.\n4. If the issue is platform-specific, first decide whether the platform is required to observe the bug or only part of the reporter's context. Stay on the current OS when it can still validate the behavior. Use a matching VM\u002Fcontainer\u002FCI path only when the repro requires that platform or when a local attempt fails and the OS mismatch remains a credible reason.\n5. If the issue has an empty body or only a design request, do not claim a repro. Turn it into a requirements\u002Fdesign investigation and identify the missing validation path.\n\nTargeted questions should ask for one missing blocker at a time, such as:\n\n| Missing blocker | Example question |\n| --- | --- |\n| Version\u002Fchannel | \"Which Aspire CLI\u002FSDK version or commit did this reproduce with?\" |\n| Platform | \"Did this happen on Windows, Linux, macOS, WSL, or inside a container?\" |\n| Project shape | \"Can you share the AppHost snippet or template\u002Foptions used to create the app?\" |\n| External service | \"Which Azure subscription\u002Fresource type or container image was used?\" |\n| UI flow | \"Which VS Code command or dashboard page interaction triggers the failure?\" |\n\n### Insufficient-information comments\n\nWhen an issue does not contain enough information to reproduce:\n\n1. Check existing comments first so you do not duplicate a recent maintainer or previous agent request.\n2. Ask only for details that unblock reproduction. Do not paste a generic checklist if one or two fields are enough.\n3. Include what you already checked so the request is credible and actionable.\n4. Present the proposed comment to the user and ask whether to post it, matching the code-review skill's \"present findings first, then ask what to do next\" pattern.\n5. Do not include sensitive logs, tokens, resource names, subscription IDs, or private URLs in the comment.\n\nAfter presenting the draft, ask for a decision with choices like:\n\n- **Post this comment** — leave the drafted comment on the issue.\n- **Edit the comment first** — revise the draft and ask again before posting.\n- **Do not comment** — skip modifying GitHub and report the investigation as blocked on missing information.\n\nOnly after the user approves posting, use a body file to avoid shell quoting issues:\n\n```bash\ncomment_file=\"$(mktemp -t aspire-issue-comment-XXXXXX.md)\"\ncat > \"$comment_file\" \u003C\u003C'EOF'\nThanks for the report. I don't have enough information to reproduce this yet.\n\nI checked the issue body and existing comments, but the repro is missing:\n\n- Aspire CLI\u002FSDK version or commit:\n- OS\u002Fshell:\n- Exact project shape or AppHost snippet:\n- Exact command\u002FUI steps that trigger the problem:\n- Expected vs actual behavior\u002Flog excerpt:\n\nCould you add those details, or attach a minimal repro project? Once we have that, we should be able to try a faithful repro.\nEOF\n\nGH_PAGER=cat PAGER=cat gh issue comment \u003Cissue-number> \\\n  --repo microsoft\u002Faspire \\\n  --body-file \"$comment_file\"\nrm \"$comment_file\"\n```\n\nTailor the missing fields to the issue area:\n\n| Area | Ask for |\n| --- | --- |\n| CLI\u002Facquisition\u002Ftemplates | `aspire --version`, install route\u002Fchannel, exact command, template\u002Foptions, clean versus existing `.aspire` state |\n| Dashboard\u002Ftelemetry | AppHost snippet, dashboard page, trace\u002Flog IDs, browser and culture, telemetry export or screenshot |\n| VS Code extension | VS Code version, extension version\u002Finstall method, configured CLI path, command\u002Fdebug flow, output window logs |\n| Integration\u002Fcomponent | Package version, resource builder code, connection string\u002Fproperties, container image or external service version |\n| App model\u002Forchestration | Resource graph\u002FAppHost snippet, endpoint ports, `aspire describe`, resource states, Docker\u002Fprocess details |\n| CI\u002Ftest | Run\u002Fjob URL, test method\u002Fproject, OS, failure log excerpt, whether it is quarantined\u002Fouterloop |\n| Deployment\u002FAzure\u002FKubernetes | Target publisher\u002Fcloud, `az`\u002F`azd` version, generated artifacts, resource group\u002Fdeployment logs, cleanup constraints |\n\n## Step 4: Prepare a faithful reproduction environment\n\nStart from a clean workspace unless the issue is explicitly about dirty state, upgrade state, or cached files.\n\nDefault to the agent's current OS and local tooling for the first repro attempt. Do not create a Windows VM, Linux container, macOS host, cloud resource, or other remote environment just because the issue mentions one in passing. Escalate to a different environment only when the issue states the behavior is OS-specific, the dependency is unavailable locally, or the current-OS attempt cannot reproduce behavior that plausibly depends on OS, shell, filesystem, networking, browser, or cloud semantics.\n\nBefore installing anything, analyze the repro steps and create a dependency readiness summary:\n\n1. List the required tools, runtimes, package managers, services, and credentials from the issue body, comments, linked repros, and logs.\n2. Check what is already available in the current environment with lightweight version probes such as `dotnet --info`, `aspire --version`, `node --version`, `npm --version`, `docker --version`, `java --version`, `az --version`, or `gh --version`.\n3. Classify each missing dependency as lightweight, heavyweight\u002Finvasive, OS-specific, cloud-only, or credential-required.\n4. Present the missing dependency summary to the user before changing the host environment when the setup is heavyweight, invasive, or ambiguous.\n\nIf a repro requires installing a large or invasive toolchain, prefer a disposable container over installing it on the host. Examples include Java\u002FPlay Framework\u002FSBT, full browser stacks, Kubernetes tooling, database servers, language runtimes not already installed, or package managers that may rewrite workspace state. For example, an issue like `#6715` that requires Java and the Play Framework should default to a containerized repro unless the user chooses local installation. Ask the user before installing heavyweight dependencies locally and offer choices like:\n\n- **Use a container (Recommended)** — create an isolated repro environment with the required tools.\n- **Install locally** — install the required software on the current machine.\n- **Do not install yet** — stop and report the repro as blocked on dependency setup.\n\nAlways include the latest stable Aspire version in the repro plan. If the issue reports an old Aspire version, first try the same repro against latest stable in the smallest faithful environment. Only install or acquire the old version when you need to confirm a historical regression, validate an upgrade path, or explain a version-specific failure. If the issue does not reproduce on latest stable and the reported version is old, offer the user an issue reply that says the current stable version was checked, the issue could not be reproduced there, and asks the reporter to upgrade or provide a latest-stable repro if it still happens.\n\n```bash\nworkdir=\"$(mktemp -d -t aspire-issue-XXXXXX)\"\ncd \"$workdir\"\n```\n\nFor repo-local investigations, restore before building from the repository root:\n\n```bash\ncd \u003Crepo-root>\n.\u002Frestore.sh\n.\u002Fbuild.sh --build \u002Fp:SkipNativeBuild=true\n```\n\nFor CLI behavior, prefer the exact reported CLI when possible:\n\n| Reported source | Repro approach |\n| --- | --- |\n| Stable\u002Fdaily\u002Fstaging script install | Run the documented install script in an isolated install path when supported |\n| PR build | Use the \"Dogfood this PR\" command and the `pr-testing` skill |\n| Local repo build | Build `src\u002FAspire.Cli\u002FAspire.Cli.csproj` and run the produced CLI via `dotnet exec` |\n| Template package | Install the reported `Aspire.ProjectTemplates` version with `dotnet new install` in an isolated temp home when practical |\n| Acquisition route bug | Preserve the route exactly: Homebrew, WinGet, MSI, npm, script, localhive, or update-in-place |\n\nFor AppHost behavior, prefer an issue-provided AppHost snippet. If none exists, create the smallest AppHost that exercises the same resource graph, endpoints, package versions, and wait relationships. Keep platform details intact: Windows shell behavior, Bun\u002Fnpm workspaces, Docker Desktop architecture, static ports, or launch profiles can be the bug.\n\nWhen running an AppHost and you need follow-up commands, prefer `start`\u002F`wait`\u002F`stop` so the app stays available while you inspect it:\n\n```bash\naspire start --project \u003Cpath-to-apphost> --non-interactive\naspire wait\naspire ps\naspire describe\naspire logs \u003Cresource> --tail 100\naspire stop\n```\n\nIf the issue involves the dashboard, use the runtime\u002Fplugin-provided `playwright-cli` skill when it is available to capture the dashboard state and reproduce the issue, including the page, telemetry export, relevant trace\u002Flog IDs, and browser console errors. If `playwright-cli` is unavailable, capture the same evidence with dashboard exports, Aspire CLI logs\u002Ftraces or runtime `aspire-monitoring` when available, screenshots, and `dashboard-testing` for test-focused dashboard issues.\n\nIf it involves cloud deployment, record generated artifacts, resource group names, `az`\u002F`azd` versions, and cleanup status.\n\n## Step 5: Reproduce and narrow the cause\n\nRun the issue's repro steps against latest stable first unless the issue is explicitly about a specific version, install\u002Fupdate route, or PR build. Then vary one factor at a time:\n\n1. Latest stable versus reported version\u002Fchannel versus current `main`.\n2. Current OS\u002Fshell versus reported OS\u002Fshell, escalating only when the mismatch plausibly explains the result.\n3. Clean cache versus existing `.aspire`, package cache, template hive, `node_modules`, or container state.\n4. Minimal AppHost versus the reported real project shape.\n5. Interactive versus `--non-interactive`.\n6. Installed CLI versus repo-local CLI.\n\nCapture the command, exit code, and relevant output for each attempt. Prefer structured evidence over screenshots when possible: logs, `aspire describe`, distributed traces, test results, workflow artifacts, generated files, or package versions.\n\nDo not stop after the first \"could not reproduce\" if the environment differs from the report in a meaningful way. Call out the mismatch and either try a closer environment or ask for the missing detail.\n\nAfter a repro attempt reaches a clear result, present the evidence to the user and ask what they want to do next instead of taking action automatically. Offer choices that match the outcome:\n\n- **Investigate root cause** — inspect the relevant code paths, tests, logs, and telemetry to explain why the reproduced behavior happens without making a fix yet.\n- **Make a fix** — do not post an issue update; implement the targeted code\u002Ftest change and open a PR once the issue is reproduced or the likely cause is strong enough.\n- **Update the issue** — draft a GitHub issue comment with the investigation evidence.\n\nBefore offering to update the issue, prepare a comment draft and scrub it for PII and sensitive details. Do not include local usernames or home directory paths, machine names, tokens, private URLs, subscription IDs, resource group names that identify a customer or user, secrets, credentials, exact private repository paths, or raw logs that may contain any of those values. Replace sensitive paths with placeholders such as `\u003Cscratch>`, `\u003Crepo>`, or `\u003Credacted>`, and summarize long logs instead of pasting them verbatim. Show the sanitized draft to the user first and ask for approval before posting.\n\n## Step 6: Investigation result format\n\nEnd every investigation with a concise status and evidence:\n\n```text\nStatus: reproduced | not reproduced | blocked | design-only | duplicate\u002Fknown\nIssue: #\u003Cnumber> \u003Ctitle>\nArea: \u003Clabels and likely code area>\nEnvironment: \u003COS, CLI\u002FSDK version, install route, runtime\u002Fcloud\u002Fbrowser details>\nRepro: \u003Cexact commands or steps that matter>\nObserved: \u003Cactual behavior with evidence>\nExpected: \u003Cexpected behavior from issue or docs>\nLikely cause: \u003Cconfirmed root cause or best hypothesis; say if unconfirmed>\nArtifacts: \u003Cpaths\u002Flinks to logs, traces, screenshots, recordings, generated files>\nProposed issue comment: \u003Csanitized draft shown to user, posted with approval, declined, or not needed>\nNext action: \u003Cask user to choose root-cause investigation, fix\u002FPR, issue update, missing info, owner question, or validation path>\n```\n\nBe explicit about confidence. Use \"confirmed\" only when the evidence directly proves the cause. Use \"hypothesis\" when the behavior is reproduced but the code-level cause is not yet verified.\n\n## Step 7: If the user also asked for a fix\n\nOnly start implementation after reproduction or a strong evidence-backed root cause. Follow the repository instructions for the touched area, add or update targeted tests, and validate with the smallest command that exercises the bug. For flaky tests, quarantined tests, CI failures, deployment E2E, dashboard tests, CLI E2E, or PR-build testing, hand off to the specialized skill instead of duplicating its workflow.\n",{"name":5,"description":7},"main","cec61ac9b795bbfd9913f0c7196689b0d7a215d7",".agents\u002Fskills\u002Fissue-investigation",1039,"https:\u002F\u002Fgithub.com\u002Fmicrosoft\u002Faspire","https:\u002F\u002Faspire.dev","https:\u002F\u002Fgithub.com\u002Fmicrosoft","https:\u002F\u002Fopensource.microsoft.com","Open source projects and samples from Microsoft","2026-11-15 00:32:23",{"source":8,"skillName":5,"commands":59,"prompt":69,"zipAvailable":70},[60,63,66],{"tool":61,"command":62},"npx","npx skills add microsoft\u002Faspire --skill issue-investigation",{"tool":64,"command":65},"bunx","bunx skills add microsoft\u002Faspire --skill issue-investigation",{"tool":67,"command":68},"pnpm","pnpm dlx skills add microsoft\u002Faspire --skill issue-investigation","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002Fmicrosoft\u002Faspire\u002Fblob\u002Fmain\u002F.agents\u002Fskills\u002Fissue-investigation\u002FSKILL.md",true,[],[73,107,131,144,162,188],{"skillNo":74,"name":75,"nameZh":6,"description":76,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":77,"categories":78,"keywords":92,"publishedAt":43},"01M6FKZQPMA7VRA0PRBZRDFJTX","vscode-extension","Use when investigating, developing, debugging, testing, or reviewing Aspire VS Code extension behavior under extension\u002F, including extension UI, command, debugger, RPC, DCP, MCP, and CLI-integration issues or features.","https:\u002F\u002Fgithub.com\u002Fmicrosoft\u002Faspire\u002Fblob\u002Fmain\u002F.agents\u002Fskills\u002Fvscode-extension\u002FSKILL.md",[79,84,87],{"categoryNo":80,"categoryName":81,"taxonomyType":82,"taxonomyTypeDesc":83},"01JBSEEDCAT000000000000008","测试与质量","SCENARIO","使用场景",{"categoryNo":85,"categoryName":86,"taxonomyType":82,"taxonomyTypeDesc":83},"01JBSEEDCAT000000000000023","Skill与智能体开发",{"categoryNo":88,"categoryName":89,"taxonomyType":90,"taxonomyTypeDesc":91},"01JBSEEDCAT000000000000030","开发工程师","PROFESSION","适用职业",[93,94,95,96,97,98,39,99,100,36,101,102,103,104,105,32,30,106,24],"vscode","vscode extension","extension","testing","MCP","cli","debugging","code","rpc","ui","integration","debugger","command","features",{"skillNo":108,"name":109,"nameZh":6,"description":110,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":111,"categories":112,"keywords":116,"publishedAt":43},"01M6FKZJG81XP80SVWFBJVMDWN","update-container-images","Updates Docker container image tags used by Aspire hosting integrations. Queries registries for newer tags, uses LLM to determine version-compatible updates, and applies changes. Use this when asked to update container image versions.","https:\u002F\u002Fgithub.com\u002Fmicrosoft\u002Faspire\u002Fblob\u002Fmain\u002F.agents\u002Fskills\u002Fupdate-container-images\u002FSKILL.md",[113],{"categoryNo":114,"categoryName":115,"taxonomyType":82,"taxonomyTypeDesc":83},"01JBSEEDCAT000000000000016","图像与多媒体",[117,118,119,120,121,122,123,39,124,30,125,126,127,128,129,25,130,24],"container images","update","container","images","update container images","image","docker","integrations","version","hosting","queries","tags","updates","changes",{"skillNo":132,"name":133,"nameZh":6,"description":134,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":135,"categories":136,"keywords":138,"publishedAt":43},"01M6FKZDY1MY0QA62EYSDG5NHW","test-management","Quarantines or disables flaky\u002Fproblematic tests using the QuarantineTools utility","https:\u002F\u002Fgithub.com\u002Fmicrosoft\u002Faspire\u002Fblob\u002Fmain\u002F.agents\u002Fskills\u002Ftest-management\u002FSKILL.md",[137],{"categoryNo":80,"categoryName":81,"taxonomyType":82,"taxonomyTypeDesc":83},[139,140,141,142,143],"test","management","test management","tests","utility",{"skillNo":145,"name":146,"nameZh":6,"description":147,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":148,"categories":149,"keywords":150,"publishedAt":43},"01M6FKZAQ428JX8F6NSA19XBQJ","startup-perf","Measures Aspire startup profiling with CLI self-profile capture and dashboard export traces.","https:\u002F\u002Fgithub.com\u002Fmicrosoft\u002Faspire\u002Fblob\u002Fmain\u002F.agents\u002Fskills\u002Fstartup-perf\u002FSKILL.md",[],[151,152,153,98,154,39,155,156,157,158,159,160,161],"perf","startup","startup perf","dashboard","profiling","capture","profile","export","self","with","traces",{"skillNo":163,"name":164,"nameZh":6,"description":165,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":166,"categories":167,"keywords":168,"publishedAt":43},"01M6FKZ4W4J0ET0Q0C2Z30HYRV","reviewing-aspire-architecture","Use only when the user explicitly requests a deep Aspire architectural or pattern review of an existing PR or diff, or when a generic reviewer escalates a named Aspire-domain question it cannot resolve. Do not use for ordinary review, implementation, debugging, explanation, or design discussion—even when a PR is referenced—and never select it from changed file paths alone.","https:\u002F\u002Fgithub.com\u002Fmicrosoft\u002Faspire\u002Fblob\u002Fmain\u002F.agents\u002Fskills\u002Freviewing-aspire-architecture\u002FSKILL.md",[],[169,39,170,171,172,173,99,174,175,176,177,30,178,179,180,181,182,183,184,185,25,186,187,24],"architecture","reviewing aspire architecture","design","review","pr","file","select","reviewer","question","diff","never","only","do","domain","pattern","deep","paths","Existing","generic",{"skillNo":189,"name":190,"nameZh":6,"description":191,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":192,"categories":193,"keywords":195,"publishedAt":43},"01M6FKYZ44BEPJNV8VQ0X73CZR","pr-testing","Use when asked to test a microsoft\u002Faspire pull request, including CLI, hosting, dashboard, component, template, and VS Code extension changes. Also covers testing PRs that change CI infrastructure — GitHub Actions (workflows, actions, gh-aw agentic workflows, CI scripts under .github\u002F and eng\u002F) and Azure DevOps pipelines (eng\u002Fpipelines, eng\u002Fcommon).","https:\u002F\u002Fgithub.com\u002Fmicrosoft\u002Faspire\u002Fblob\u002Fmain\u002F.agents\u002Fskills\u002Fpr-testing\u002FSKILL.md",[194],{"categoryNo":80,"categoryName":81,"taxonomyType":82,"taxonomyTypeDesc":83},[96,173,196,98,33,197,154,198,199,200,201,202,39,203,100,204,205,206,207,208,30,209,126,210,95,211,130,212],"pr testing","agentic","workflows","change","component","agentic workflows","azure devops","infrastructure","actions","ci","github actions","template","common","devops","pipelines","scripts","prs"]