feat: Add GitHub issue automation with issue metadata #350
There are no files selected for viewing
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
| Original file line number | Diff line number | Diff line change |
|---|---|---|
| @@ -0,0 +1,165 @@ | ||
| name: 'Strands Task Labeller' | ||
| description: 'Automatically analyze and label GitHub issues using Strands agent' | ||
| inputs: | ||
| aws_role_arn: | ||
| description: 'AWS IAM role ARN for authentication' | ||
| required: true | ||
| sessions_bucket: | ||
| description: 'S3 bucket for session storage' | ||
| required: true | ||
| pat_token: | ||
| description: 'pat token to modify repo' | ||
| required: false | ||
|
|
||
| runs: | ||
| using: 'composite' | ||
| steps: | ||
| - name: Checkout repository | ||
| uses: actions/checkout@v6 | ||
| with: | ||
| sparse-checkout: | | ||
| .github | ||
|
|
||
| - name: Copy .github to safe directory | ||
| shell: bash | ||
| run: | | ||
| mkdir -p ${{ runner.temp }}/strands-labeller | ||
| cp -r .github ${{ runner.temp }}/strands-labeller | ||
|
|
||
| - name: Set up Python | ||
| uses: actions/setup-python@v4 | ||
| with: | ||
| python-version: '3.13' | ||
|
|
||
| - name: Install uv | ||
| uses: astral-sh/setup-uv@v3 | ||
| with: | ||
| enable-cache: true | ||
| cache-dependency-glob: '${{ runner.temp }}/strands-labeller/.github/scripts/python/requirements.txt' | ||
|
|
||
| - name: Install Strands Agents | ||
| shell: bash | ||
| run: | | ||
| echo "📦 Installing from requirements.txt" | ||
| uv pip install --system -r ${{ runner.temp }}/strands-labeller/.github/scripts/python/requirements.txt --quiet | ||
|
|
||
| - name: Configure Git | ||
| shell: bash | ||
| run: | | ||
| git config --global user.name "Strands Agent" | ||
| git config --global user.email "217235299+strands-agent@users.noreply.github.com" | ||
| git config --global core.pager cat | ||
| PAGER=cat | ||
|
|
||
| - name: Configure AWS credentials | ||
| uses: aws-actions/configure-aws-credentials@v4 | ||
| with: | ||
| role-to-assume: ${{ inputs.aws_role_arn }} | ||
| role-session-name: GitHubActions-StrandsAgent-${{ github.run_id }} | ||
| aws-region: us-west-2 | ||
| mask-aws-account-id: true | ||
| inline-session-policy: >- | ||
| { | ||
| "Version": "2012-10-17", | ||
| "Statement": [ | ||
| { | ||
| "Sid":"BedrockAccess", | ||
| "Effect": "Allow", | ||
| "Action": [ | ||
| "bedrock:InvokeModelWithResponseStream", | ||
| "bedrock:InvokeModel" | ||
| ], | ||
| "Resource": "*" | ||
| }, { | ||
| "Effect": "Allow", | ||
| "Action": [ | ||
| "s3:PutObject", | ||
| "s3:GetObject", | ||
| "s3:DeleteObject" | ||
| ], | ||
| "Resource": [ | ||
| "arn:aws:s3:::${{ inputs.sessions_bucket }}/*" | ||
| ] | ||
| }, { | ||
| "Effect": "Allow", | ||
| "Action": "s3:ListBucket", | ||
| "Resource": [ | ||
| "arn:aws:s3:::${{ inputs.sessions_bucket }}" | ||
| ] | ||
| } | ||
| ] | ||
| } | ||
|
|
||
| - name: Execute labelling task | ||
| shell: bash | ||
| env: | ||
| GITHUB_TOKEN: ${{ github.token }} | ||
| PAT_TOKEN: ${{ inputs.pat_token }} | ||
| GITHUB_REPOSITORY: ${{ github.repository }} | ||
| GITHUB_WRITE: 'true' | ||
| S3_SESSION_BUCKET: ${{ inputs.sessions_bucket }} | ||
| SESSION_ID: 'labeller-${{ github.event.issue.number }}-${{ github.run_id }}' | ||
| INPUT_SYSTEM_PROMPT: 'You are a GitHub Issue Categorization Agent. Your role is to analyze new GitHub issues and automatically apply appropriate labels and tag relevant area experts. | ||
|
|
||
| ## Core Responsibilities | ||
| 1. **Label Assignment**: Apply relevant labels from the repository based on issue content | ||
| 2. **Area Expert Tagging**: Tag appropriate team members using @mentions based on technical areas | ||
| 3. **Issue Summary**: Generate a brief summary with context from similar issues | ||
|
|
||
| ## Process | ||
| 1. First, search for similar issues using search_similar_issues tool to understand context | ||
| 2. Get available repository labels using list_repository_labels tool | ||
| 3. Read the automation configuration from .github/config/automation-config.json for area experts | ||
| 4. Analyze the issue title, description, and any provided context | ||
| 5. Apply ONLY existing labels from step 2 using add_labels_to_issue tool | ||
| 6. Tag relevant area experts with @mentions in a comment | ||
|
|
||
| ## Similar Issues Analysis | ||
| - Use search_similar_issues tool to find related issues before categorizing | ||
| - Extract keywords from the new issue title and description | ||
| - Review similar issues to understand patterns and common solutions | ||
| - Reference similar issues in your summary if relevant | ||
|
|
||
| ## Label Assignment | ||
| - CRITICAL: Use list_repository_labels tool FIRST to get available labels | ||
| - ONLY apply labels that exist in the repository - NEVER create new labels | ||
| - Match issue content to existing labels (bug, enhancement, documentation, etc.) | ||
| - If no perfect match exists, choose the closest existing label | ||
| - Focus on categorization labels that help with organization | ||
|
|
||
| ## Area Expert Assignment Rules | ||
| - Match issue content to technical areas (typescript, bedrock, openai, mcp, etc.) | ||
| - Tag ONLY the 2 most relevant experts based on the area_experts configuration | ||
| - Use @username format for mentions | ||
| - Prioritize experts most closely related to the specific issue content | ||
|
|
||
| ## Area Expert Matching | ||
| Match issue content to these technical areas and tag corresponding experts: | ||
| - agent_loop, context_management, async_streaming, tool_executors | ||
| - bi_directional_streaming, human_in_loop, hooks | ||
| - bedrock, anthropic, gemini, litellm, llamaapi, llama_cpp, mistralai, ollama, openai, sagemaker, writer, cohere | ||
| - multi_agent, a2a, sessions, memory, mcp, structured_outputs | ||
| - agentcore_integrations, telemetry, evaluations, github_workflows | ||
| - website_docs, tools, samples, agent_builder, mcp_server | ||
| - embodied_ai, typescript, constraints_engine | ||
|
|
||
| ## Output Format | ||
| - Apply labels immediately using add_labels_to_issue tool | ||
| - Post a single comment with expert mentions that includes: | ||
| - Brief issue summary with context from similar issues (if found) | ||
| - Expert mentions (MAX 2 people): "CC @expert1 @expert2 for [specific area] - [brief reason why they are needed]" | ||
| - Links to similar issues if relevant | ||
| - Clear action items or next steps if applicable | ||
| - Keep comments professional but informative | ||
| - For TypeScript issues: Only tag TypeScript experts if the issue is specifically about TypeScript compilation, syntax, or type errors (not just because this is a TypeScript project) | ||
|
|
||
| ## Important Notes | ||
| - Do NOT set any project fields or priorities - this agent is for categorization only | ||
| - Focus on helping organize and route issues to the right people | ||
| - Prioritization will be handled by a separate process | ||
|
|
||
| Use GitHub tools to execute all actions. Be thorough but efficient in your analysis.' | ||
| STRANDS_TOOL_CONSOLE_MODE: 'enabled' | ||
| BYPASS_TOOL_CONSENT: 'true' | ||
JackYPCOnline marked this conversation as resolved.
Show resolved
Hide resolved
|
||
| run: | | ||
| uv run ${{ runner.temp }}/strands-labeller/.github/scripts/python/agent_runner.py "Analyze and categorize GitHub issue #${{ github.event.issue.number }}: search for similar issues, apply appropriate labels, and tag relevant area experts based on issue content." | ||
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
| Original file line number | Diff line number | Diff line change |
|---|---|---|
| @@ -0,0 +1,148 @@ | ||
| { | ||
| "area_experts": { | ||
| "agent_loop": { | ||
| "name": "Agent Loop", | ||
| "github_usernames": ["pgrayy", "zastrowm"] | ||
| }, | ||
| "context_management": { | ||
| "name": "Context Management / Windows", | ||
| "github_usernames": ["Unshure", "afarntrog"] | ||
| }, | ||
| "async_streaming": { | ||
| "name": "Async / Callbacks Streaming", | ||
| "github_usernames": ["pgrayy"] | ||
| }, | ||
| "tool_executors": { | ||
| "name": "Tool Executors", | ||
| "github_usernames": ["pgrayy"] | ||
| }, | ||
| "bi_directional_streaming": { | ||
| "name": "Bi-directional Streaming", | ||
| "github_usernames": ["mehtarac", "mkmeral", "pgrayy"] | ||
| }, | ||
| "human_in_loop": { | ||
| "name": "Human in the Loop / Interrupts", | ||
| "github_usernames": ["mkmeral", "pgrayy"] | ||
| }, | ||
| "hooks": { | ||
| "name": "Hooks", | ||
| "github_usernames": ["zastrowm"] | ||
| }, | ||
| "bedrock": { | ||
| "name": "Bedrock Model Provider", | ||
| "github_usernames": ["mehtarac"] | ||
| }, | ||
| "anthropic": { | ||
| "name": "Anthropic Model Provider", | ||
| "github_usernames": ["pgrayy"] | ||
| }, | ||
| "gemini": { | ||
| "name": "Gemini Model Provider", | ||
| "github_usernames": ["notgitika"] | ||
| }, | ||
| "litellm": { | ||
|
Member
There was a problem hiding this comment. Would it be better to list the the label associated with the category? Or are we allowing any label? Right now all of our "area" label start with "area-"
Author
There was a problem hiding this comment. For labels, the agent will only use existing labels. It will @ the person in this list in the comment, thus the expert will receive an email noticification. We can do assign the expert directly, but personally I think this is a little aggressive.
Member
There was a problem hiding this comment. Agreed, we shouldn't assign. I like mentioning more. But we do have to be concerned with user-submitted content possibly spamming the repository via the agent. Some sort of guard rails around what the agent can do I think would be ideal. |
||
| "name": "LiteLLM Model Provider", | ||
| "github_usernames": ["dbschmigelski"] | ||
| }, | ||
| "llamaapi": { | ||
| "name": "LlamaAPI Model Provider", | ||
| "github_usernames": ["pgrayy"] | ||
| }, | ||
| "llama_cpp": { | ||
| "name": "llama.cpp Model Provider", | ||
| "github_usernames": ["awsarron"] | ||
| }, | ||
| "mistralai": { | ||
| "name": "MistralAI Model Provider", | ||
| "github_usernames": ["mehtarac"] | ||
| }, | ||
| "ollama": { | ||
| "name": "Ollama Model Provider", | ||
| "github_usernames": ["dbschmigelski"] | ||
| }, | ||
| "openai": { | ||
| "name": "OpenAI Model Provider", | ||
| "github_usernames": ["pgrayy"] | ||
| }, | ||
| "sagemaker": { | ||
| "name": "SageMaker Model Provider", | ||
| "github_usernames": ["mehtarac"] | ||
| }, | ||
| "writer": { | ||
| "name": "Writer Model Provider", | ||
| "github_usernames": ["mehtarac"] | ||
| }, | ||
| "cohere": { | ||
| "name": "Cohere Model Provider", | ||
| "github_usernames": ["Unshure"] | ||
| }, | ||
| "multi_agent": { | ||
| "name": "Multi-agent Orchestrators", | ||
| "github_usernames": ["mkmeral", "JackYPCOnline", "awsarron"] | ||
| }, | ||
| "a2a": { | ||
| "name": "A2A", | ||
| "github_usernames": ["awsarron"] | ||
| }, | ||
| "sessions": { | ||
| "name": "Sessions", | ||
| "github_usernames": ["Unshure", "JackYPCOnline"] | ||
| }, | ||
| "memory": { | ||
| "name": "Memory", | ||
| "github_usernames": ["Unshure", "mehtarac", "afarntrog"] | ||
| }, | ||
| "mcp": { | ||
| "name": "MCP", | ||
| "github_usernames": ["dbschmigelski"] | ||
| }, | ||
| "structured_outputs": { | ||
| "name": "Structured Outputs", | ||
| "github_usernames": ["afarntrog"] | ||
| }, | ||
| "agentcore_integrations": { | ||
| "name": "AgentCore Integrations", | ||
| "github_usernames": ["JackYPCOnline", "afarntrog", "mehtarac", "dbschmigelski"] | ||
| }, | ||
| "telemetry": { | ||
| "name": "Telemetry / Observability", | ||
| "github_usernames": ["poshinchen", "JackYPCOnline"] | ||
| }, | ||
| "evaluations": { | ||
| "name": "Evaluations", | ||
| "github_usernames": ["poshinchen"] | ||
| }, | ||
| "github_workflows": { | ||
| "name": "GitHub Workflows and Actions", | ||
| "github_usernames": ["dbschmigelski", "yonib05", "Unshure"] | ||
| }, | ||
| "website_docs": { | ||
| "name": "Website and Docs", | ||
| "github_usernames": ["zastrowm", "yonib05", "rycolez"] | ||
| }, | ||
| "tools": { | ||
| "name": "Tools", | ||
| "github_usernames": ["cagataycali", "JackYPCOnline", "zastrowm", "mehtarac"] | ||
| }, | ||
| "agent_builder": { | ||
| "name": "Agent Builder", | ||
| "github_usernames": ["cagataycali"] | ||
| }, | ||
| "mcp_server": { | ||
| "name": "MCP Server", | ||
| "github_usernames": ["JackYPCOnline", "mkmeral"] | ||
| }, | ||
| "embodied_ai": { | ||
| "name": "Embodied / Physical AI and Edge Inference / Robotics", | ||
| "github_usernames": ["awsarron", "cagataycali", "mkmeral"] | ||
| }, | ||
| "typescript": { | ||
| "name": "TypeScript", | ||
| "github_usernames": ["Unshure", "zastrowm", "afarntrog", "awsarron"] | ||
| }, | ||
| "steering": { | ||
| "name": "Steering", | ||
| "github_usernames": ["dbschmigelski"] | ||
| } | ||
| } | ||
| } | ||
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Oops, something went wrong.
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Uh oh!
There was an error while loading. Please reload this page.