Merge branch 'main' into copilot/add-test-splitting-infrastructure-again

Author: radical

.github/copilot-instructions.md

@@ -183,137 +183,6 @@ The `*.Designer.cs` files are in the repo, but are intended to match same named
 * Code blocks should be formatted with triple backticks (```) and include the language identifier for syntax highlighting.
 * JSON code blocks should be indented properly.
 
-## Custom Skills System
-
-This repository supports custom skills - reusable, specialized capabilities that extend Copilot's functionality for specific tasks. Skills work like plugins that you should automatically invoke when relevant to the user's request.
-
-### Skill Locations
-
-Skills are stored in the following directories (in order of precedence):
-1. **Project Skills** (shared with team): `.claude/skills/` - checked into git
-2. **Personal Skills** (user-specific): `~/.claude/skills/` - not checked into git
-
-### Skill Structure
-
-Each skill is a directory containing a `SKILL.md` file with this format:
-
-```markdown
----
-name: Human Readable Name (max 64 chars)
-description: One-line description of what the skill does and when to use it (max 1024 chars)
-allowed-tools: Read, Grep, Glob, Bash, Edit, Write  # Optional: restrict which tools the skill can use
----
-
-# Skill Name
-
-## Purpose
-Clear explanation of what this skill does.
-
-## When to Use
-Specific scenarios when this skill should be invoked.
-
-## Instructions
-Step-by-step guidance for executing the skill.
-
-## Examples
-Concrete examples demonstrating the skill's usage.
-```
-
-### How to Discover and Use Skills
-
-**IMPORTANT**: You must proactively check for and use skills when they match the user's request.
-
-#### Step 1: Skill Discovery
-
-When a user makes a request, before starting work:
-
-1. **Search for relevant skills** by checking `.claude/skills/` and `~/.claude/skills/`:
-   ```bash
-   # List all available skills
-   find .claude/skills -name "SKILL.md" -type f 2>/dev/null
-   find ~/.claude/skills -name "SKILL.md" -type f 2>/dev/null
-   ```
-
-2. **Read skill descriptions** to find matches:
-   - Read the YAML frontmatter of each `SKILL.md` file
-   - Match the `description` field against the user's request
-   - A skill is relevant if its description keywords overlap with the task
-
-#### Step 2: Skill Invocation
-
-When you identify a relevant skill:
-
-1. **Announce skill activation** to the user:
-   ```
-   I found a relevant skill: "{skill_name}". I'll use this to help with your request.
-   ```
-
-2. **Read the complete SKILL.md file** to understand the full instructions:
-   - Use the Read tool to load the entire skill file
-   - Parse both the frontmatter and the markdown content
-   - Pay special attention to the Instructions and Examples sections
-
-3. **Follow the skill's instructions exactly**:
-   - Execute each step in the Instructions section
-   - Use only the tools specified in `allowed-tools` (if defined)
-   - Adapt the examples to the current context
-   - If the skill references other files in its directory, read them as needed
-
-4. **Report completion**:
-   ```
-   Completed task using the "{skill_name}" skill.
-   ```
-
-#### Step 3: When to Use Subagents for Skills
-
-For complex skills that involve multiple phases or extensive codebase exploration, use subagents:
-
-1. **Use Task tool with subagent_type=Explore** when a skill requires:
-   - Searching across multiple files or directories
-   - Understanding codebase patterns before making changes
-   - Gathering context from various locations
-
-2. **Use Task tool with subagent_type=general-purpose** when a skill requires:
-   - Multi-step workflows that could fail and need retry logic
-   - Complex decision trees based on what's found in the codebase
-   - Long-running operations that benefit from isolation
-
-Example of using a subagent for skill execution:
-```
-I'll use the Task tool to invoke the "{skill_name}" skill:
-[Invoke Task tool with subagent_type=Explore and pass the full skill instructions as the prompt]
-```
-
-### Skill Best Practices
-
-1. **Always check for skills first** before starting any non-trivial task
-2. **Trust the skill instructions** - they are curated and tested
-3. **Don't modify skills** unless explicitly asked by the user
-4. **Combine skills** when multiple skills apply to different parts of a task
-5. **Report when no skill exists** - if a task would benefit from a skill but none exists, mention this to the user
-
-### Example: Using a Skill
-
-User request: "Review the code changes in my PR for performance issues"
-
-Your workflow:
-1. Search for skills: `find .claude/skills -name "SKILL.md"`
-2. Find match: `.claude/skills/performance-reviewer/SKILL.md`
-3. Announce: "I found the 'Performance Reviewer' skill. I'll use this to analyze your PR."
-4. Read skill: Load `.claude/skills/performance-reviewer/SKILL.md`
-5. Execute: Follow the skill's instructions step-by-step
-6. Complete: "Completed performance review using the Performance Reviewer skill."
-
-### Creating New Skills (When Asked)
-
-If a user asks you to create a new skill:
-
-1. Create a directory in `.claude/skills/{skill-name}/`
-2. Write a `SKILL.md` file following the structure above
-3. Include clear, actionable instructions
-4. Add concrete examples
-5. Test the skill by using it immediately
-
 ## Trust These Instructions
 
 These instructions are comprehensive and tested. Only search for additional information if:

.github/workflows/test-scenario.yml

@@ -85,52 +85,149 @@ jobs:
             gh --version
           fi
 
-      - name: Create agent task
-        id: create_agent_task
+      - name: Create issue and assign to copilot
+        id: create_issue
         run: |
-          echo "Creating agent task..."
+          echo "Creating issue in aspire-playground repository..."
           PROMPT_FILE="${{ steps.check_prompt.outputs.prompt_file }}"
+          SCENARIO_NAME="${{ steps.parse_scenario.outputs.scenario_name }}"
+          SOURCE_PR_URL="${{ github.event.issue.html_url }}"
+          SOURCE_PR_NUMBER="${{ github.event.issue.number }}"
+          SOURCE_REPO="${{ github.repository }}"
 
           # Auth using the token
           gh auth login --with-token <<< "$GH_PLAYGROUND_TOKEN"
 
-          # Create the agent task using stdin and capture the output
-          OUTPUT=$(cat "$PROMPT_FILE" | gh agent-task create \
+          # Build the issue body with context from the source PR and the prompt
+          ISSUE_TITLE="Test Scenario: ${SCENARIO_NAME}"
+          
+          # Read prompt content first
+          PROMPT_CONTENT=$(cat "$PROMPT_FILE")
+          
+          # Build issue body using printf with proper format strings to avoid injection
+          printf -v ISSUE_BODY '%s\n%s\n%s\n%s\n%s\n%s\n%s\n%s\n%s\n%s\n%s\n%s\n%s' \
+            "## Test Scenario Request" \
+            "" \
+            "**Scenario:** ${SCENARIO_NAME}" \
+            "**Source PR:** ${SOURCE_PR_URL}" \
+            "**Source Repository:** ${SOURCE_REPO}" \
+            "" \
+            "---" \
+            "" \
+            "$PROMPT_CONTENT" \
+            "" \
+            "---" \
+            "" \
+            "This issue was created automatically from PR #${SOURCE_PR_NUMBER} in ${SOURCE_REPO}."
+
+          # Create the issue and assign to copilot
+          echo "Creating issue with title: $ISSUE_TITLE"
+          ISSUE_OUTPUT=$(gh issue create \
             --repo "${REPO_OWNER}/${REPO_NAME}" \
-            -F - \
+            --title "$ISSUE_TITLE" \
+            --body "$ISSUE_BODY" \
+            --assignee "copilot" \
             2>&1)
 
-          echo "Agent task output:"
-          echo "$OUTPUT"
+          echo "Issue creation output:"
+          echo "$ISSUE_OUTPUT"
 
-          # Extract the PR URL from the output
-          PR_URL=$(echo "$OUTPUT" | \
-            grep -oP 'https://github.com/[^/]+/[^/]+/pull/\d+' | head -1)
+          # Extract the issue URL from the output
+          ISSUE_URL=$(echo "$ISSUE_OUTPUT" | \
+            grep -oP 'https://github.com/[^/]+/[^/]+/issues/\d+' | head -1)
 
-          if [ -z "$PR_URL" ]; then
-            echo "Warning: Could not extract PR URL from output"
-            echo "pr_url=" >> $GITHUB_OUTPUT
+          if [ -z "$ISSUE_URL" ]; then
+            echo "Error: Could not extract issue URL from output"
+            exit 1
+          fi
+
+          echo "Successfully created issue: $ISSUE_URL"
+          echo "issue_url=$ISSUE_URL" >> $GITHUB_OUTPUT
+
+          # Extract issue number for later use
+          ISSUE_NUMBER=$(echo "$ISSUE_URL" | grep -oP '/issues/\K\d+')
+          echo "issue_number=$ISSUE_NUMBER" >> $GITHUB_OUTPUT
+
+      - name: Wait for agent PR to be created
+        id: wait_for_pr
+        run: |
+          echo "Waiting for GitHub Copilot agent to create a PR..."
+          ISSUE_NUMBER="${{ steps.create_issue.outputs.issue_number }}"
+          MAX_ATTEMPTS=30
+          SLEEP_SECONDS=10
+          ATTEMPT=0
+          FOUND_PR=""
+
+          # Auth using the token
+          gh auth login --with-token <<< "$GH_PLAYGROUND_TOKEN"
+
+          while [ $ATTEMPT -lt $MAX_ATTEMPTS ]; do
+            ATTEMPT=$((ATTEMPT + 1))
+            echo "Attempt $ATTEMPT of $MAX_ATTEMPTS..."
+
+            # Try to find linked PRs by searching for PRs that reference this issue
+            LINKED_PRS=$(gh pr list \
+              --repo "${REPO_OWNER}/${REPO_NAME}" \
+              --search "linked:issue-${ISSUE_NUMBER}" \
+              --json number,url,title \
+              --jq '.[0].url' 2>/dev/null || echo "")
+
+            if [ -n "$LINKED_PRS" ] && [ "$LINKED_PRS" != "null" ] && [ "$LINKED_PRS" != "" ]; then
+              echo "Found linked PR: $LINKED_PRS"
+              FOUND_PR="$LINKED_PRS"
+              break
+            fi
+
+            # Also check for PRs created by the copilot bot recently
+            COPILOT_PRS=$(gh pr list \
+              --repo "${REPO_OWNER}/${REPO_NAME}" \
+              --author "copilot" \
+              --limit 5 \
+              --json number,url,title,createdAt \
+              --jq 'sort_by(.createdAt) | reverse | .[0].url' 2>/dev/null || echo "")
+
+            if [ -n "$COPILOT_PRS" ] && [ "$COPILOT_PRS" != "null" ] && [ "$COPILOT_PRS" != "" ]; then
+              echo "Found recent copilot PR: $COPILOT_PRS"
+              FOUND_PR="$COPILOT_PRS"
+              break
+            fi
+
+            if [ $ATTEMPT -lt $MAX_ATTEMPTS ]; then
+              echo "No PR found yet, waiting ${SLEEP_SECONDS} seconds..."
+              sleep $SLEEP_SECONDS
+            fi
+          done
+
+          if [ -n "$FOUND_PR" ]; then
+            echo "pr_url=$FOUND_PR" >> $GITHUB_OUTPUT
           else
-            echo "Successfully created agent task with PR: $PR_URL"
-            echo "pr_url=$PR_URL" >> $GITHUB_OUTPUT
+            echo "Warning: No PR found after $MAX_ATTEMPTS attempts"
+            echo "The agent may still be working on it."
+            echo "pr_url=" >> $GITHUB_OUTPUT
           fi
 
-      - name: Comment on PR with agent task link
-        if: steps.create_agent_task.outputs.pr_url != ''
+      - name: Comment on PR with issue and agent PR links
         uses: actions/github-script@60a0d83039c74a4aee543508d2ffcb1c3799cdea  # v7.0.1
         with:
           github-token: ${{ secrets.GITHUB_TOKEN }}
           script: |
-            const prUrl = '${{ steps.create_agent_task.outputs.pr_url }}';
+            const issueUrl = '${{ steps.create_issue.outputs.issue_url }}';
+            const prUrl = '${{ steps.wait_for_pr.outputs.pr_url }}';
             const scenarioName = '${{ steps.parse_scenario.outputs.scenario_name }}';
-            const comment = `🤖 **AI Agent Task Created**
+
+            let comment = `🤖 **AI Agent Task Created**
 
             Scenario: **${scenarioName}**
 
-            An AI agent has been triggered to execute this scenario.
-            You can track the progress here:
+            An AI agent has been assigned to execute this scenario.
+
+            📝 **Issue:** ${issueUrl}`;
 
-            ${prUrl}`;
+            if (prUrl) {
+              comment += `\n🔀 **Agent PR:** ${prUrl}`;
+            } else {
+              comment += `\n\n⏳ The agent is working on this task. The PR will be linked in the issue once created.`;
+            }
 
             await github.rest.issues.createComment({
               issue_number: context.issue.number,

Aspire.slnx

@@ -26,7 +26,7 @@
     <Project Path="src/Components/Aspire.Microsoft.EntityFrameworkCore.SqlServer/Aspire.Microsoft.EntityFrameworkCore.SqlServer.csproj" />
     <Project Path="src/Components/Aspire.Microsoft.Extensions.Configuration.AzureAppConfiguration/Aspire.Microsoft.Extensions.Configuration.AzureAppConfiguration.csproj" />
     <Project Path="src/Components/Aspire.Milvus.Client/Aspire.Milvus.Client.csproj" />
-    <Project Path="src/Components/Aspire.MongoDB.Driver.v3/Aspire.MongoDB.Driver.v3.csproj" />
+    <Project Path="src/Components/Aspire.MongoDB.Driver.v2/Aspire.MongoDB.Driver.v2.csproj" />
     <Project Path="src/Components/Aspire.MongoDB.Driver/Aspire.MongoDB.Driver.csproj" />
     <Project Path="src/Components/Aspire.MySqlConnector/Aspire.MySqlConnector.csproj" />
     <Project Path="src/Components/Aspire.NATS.Net/Aspire.NATS.Net.csproj" />
@@ -36,7 +36,7 @@
     <Project Path="src/Components/Aspire.Oracle.EntityFrameworkCore/Aspire.Oracle.EntityFrameworkCore.csproj" />
     <Project Path="src/Components/Aspire.Pomelo.EntityFrameworkCore.MySql/Aspire.Pomelo.EntityFrameworkCore.MySql.csproj" />
     <Project Path="src/Components/Aspire.Qdrant.Client/Aspire.Qdrant.Client.csproj" />
-    <Project Path="src/Components/Aspire.RabbitMQ.Client.v7/Aspire.RabbitMQ.Client.v7.csproj" />
+    <Project Path="src/Components/Aspire.RabbitMQ.Client.v6/Aspire.RabbitMQ.Client.v6.csproj" />
     <Project Path="src/Components/Aspire.RabbitMQ.Client/Aspire.RabbitMQ.Client.csproj" />
     <Project Path="src/Components/Aspire.Seq/Aspire.Seq.csproj" />
     <Project Path="src/Components/Aspire.StackExchange.Redis.DistributedCaching/Aspire.StackExchange.Redis.DistributedCaching.csproj" />
@@ -106,6 +106,11 @@
     <File Path="playground/README.md" />
     <Project Path="playground/Playground.ServiceDefaults/Playground.ServiceDefaults.csproj" />
   </Folder>
+  <Folder Name="/playground/AspireWithJavaScript/">
+    <Project Path="playground/AspireWithJavaScript/AspireJavaScript.AppHost/AspireJavaScript.AppHost.csproj" />
+    <Project Path="playground/AspireWithJavaScript/AspireJavaScript.MinimalApi/AspireJavaScript.MinimalApi.csproj" />
+    <Project Path="playground/AspireWithJavaScript/AspireJavaScript.ServiceDefaults/AspireJavaScript.ServiceDefaults.csproj" />
+  </Folder>
   <Folder Name="/playground/AzureAIFoundryEndToEnd/">
     <Project Path="playground/AzureAIFoundryEndToEnd/AzureAIFoundryEndToEnd.AppHost/AzureAIFoundryEndToEnd.AppHost.csproj" />
     <Project Path="playground/AzureAIFoundryEndToEnd/AzureAIFoundryEndToEnd.WebStory/AzureAIFoundryEndToEnd.WebStory.csproj" />
@@ -176,8 +181,8 @@
     <Project Path="playground/deployers/Deployers.AppHost/Deployers.AppHost.csproj" />
   </Folder>
   <Folder Name="/playground/DevTunnels/">
-    <Project Path="playground/DevTunnels/DevTunnels.AppHost/DevTunnels.AppHost.csproj" />
     <Project Path="playground/DevTunnels/DevTunnels.ApiService/DevTunnels.ApiService.csproj" />
+    <Project Path="playground/DevTunnels/DevTunnels.AppHost/DevTunnels.AppHost.csproj" />
     <Project Path="playground/DevTunnels/DevTunnels.WebFrontEnd/DevTunnels.WebFrontEnd.csproj" />
   </Folder>
   <Folder Name="/playground/dockerfile/">
@@ -199,11 +204,6 @@
   <Folder Name="/playground/HealthChecks/">
     <Project Path="playground/HealthChecks/HealthChecksSandbox.AppHost/HealthChecksSandbox.AppHost.csproj" />
   </Folder>
-  <Folder Name="/playground/javascript/">
-    <Project Path="playground/AspireWithJavaScript/AspireJavaScript.AppHost/AspireJavaScript.AppHost.csproj" />
-    <Project Path="playground/AspireWithJavaScript/AspireJavaScript.MinimalApi/AspireJavaScript.MinimalApi.csproj" />
-    <Project Path="playground/AspireWithJavaScript/AspireJavaScript.ServiceDefaults/AspireJavaScript.ServiceDefaults.csproj" />
-  </Folder>
   <Folder Name="/playground/kafka/">
     <Project Path="playground/kafka/Consumer/Consumer.csproj" />
     <Project Path="playground/kafka/KafkaBasic.AppHost/KafkaBasic.AppHost.csproj" />
@@ -234,8 +234,8 @@
   </Folder>
   <Folder Name="/playground/node/">
     <Project Path="playground/AspireWithNode/AspireWithNode.AppHost/AspireWithNode.AppHost.csproj" />
-    <Project Path="playground/AspireWithNode/AspireWithNode.ServiceDefaults/AspireWithNode.ServiceDefaults.csproj" />
     <Project Path="playground/AspireWithNode/AspireWithNode.AspNetCoreApi/AspireWithNode.AspNetCoreApi.csproj" />
+    <Project Path="playground/AspireWithNode/AspireWithNode.ServiceDefaults/AspireWithNode.ServiceDefaults.csproj" />
   </Folder>
   <Folder Name="/playground/OpenAIEndToEnd/">
     <Project Path="playground/OpenAIEndToEnd/OpenAIEndToEnd.AppHost/OpenAIEndToEnd.AppHost.csproj" />
@@ -392,7 +392,7 @@
     <Project Path="tests/Aspire.Microsoft.Extensions.Configuration.AzureAppConfiguration.Tests/Aspire.Microsoft.Extensions.Configuration.AzureAppConfiguration.Tests.csproj" />
     <Project Path="tests/Aspire.Milvus.Client.Tests/Aspire.Milvus.Client.Tests.csproj" />
     <Project Path="tests/Aspire.MongoDB.Driver.Tests/Aspire.MongoDB.Driver.Tests.csproj" />
-    <Project Path="tests/Aspire.MongoDB.Driver.v3.Tests/Aspire.MongoDB.Driver.v3.Tests.csproj" />
+    <Project Path="tests/Aspire.MongoDB.Driver.v2.Tests/Aspire.MongoDB.Driver.v2.Tests.csproj" />
     <Project Path="tests/Aspire.MySqlConnector.Tests/Aspire.MySqlConnector.Tests.csproj" />
     <Project Path="tests/Aspire.NATS.Net.Tests/Aspire.NATS.Net.Tests.csproj" />
     <Project Path="tests/Aspire.Npgsql.EntityFrameworkCore.PostgreSQL.Tests/Aspire.Npgsql.EntityFrameworkCore.PostgreSQL.Tests.csproj" />
@@ -402,7 +402,7 @@
     <Project Path="tests/Aspire.Pomelo.EntityFrameworkCore.MySql.Tests/Aspire.Pomelo.EntityFrameworkCore.MySql.Tests.csproj" />
     <Project Path="tests/Aspire.Qdrant.Client.Tests/Aspire.Qdrant.Client.Tests.csproj" />
     <Project Path="tests/Aspire.RabbitMQ.Client.Tests/Aspire.RabbitMQ.Client.Tests.csproj" />
-    <Project Path="tests/Aspire.RabbitMQ.Client.v7.Tests/Aspire.RabbitMQ.Client.v7.Tests.csproj" />
+    <Project Path="tests/Aspire.RabbitMQ.Client.v6.Tests/Aspire.RabbitMQ.Client.v6.Tests.csproj" />
     <Project Path="tests/Aspire.Seq.Tests/Aspire.Seq.Tests.csproj" />
     <Project Path="tests/Aspire.StackExchange.Redis.DistributedCaching.Tests/Aspire.StackExchange.Redis.DistributedCaching.Tests.csproj" />
     <Project Path="tests/Aspire.StackExchange.Redis.OutputCaching.Tests/Aspire.StackExchange.Redis.OutputCaching.Tests.csproj" />