sdk main #7

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Draft

dbschmigelski wants to merge 47 commits into main from sdk-main

.github/actions/README.md

-Original file line number
+Diff line change
@@ Expand Up @@
 . **Create S3 Bucket** for session storage
 . **Add GitHub Secrets**:
        - `AWS_ROLE_ARN`: The created role ARN
-       - `STRANDS_SESSION_BUCKET`: The S3 bucket name
+       - `AGENT_SESSIONS_BUCKET`: The S3 bucket name
     ## Security
@@ Expand Down @@

.github/actions/strands-agent-runner/action.yml

-Original file line number
+Diff line change
@@ Expand Up / @@ -149,7 +149,7 @@ runs: @@
             STRANDS_TOOL_CONSOLE_MODE: 'enabled'
             BYPASS_TOOL_CONSENT: 'true'
           run: |
-            uv run --no-project ${{ runner.temp }}/strands-agent-runner/.github/scripts/python/agent_runner.py "$INPUT_TASK"
+            uv run --no-project ${{ runner.temp }}/strands-agent-runner/.github/scripts/python/agent_runner.py
         - name: Capture repository state
           shell: bash
@@ Expand Down @@

.github/agent-sops/task-release-notes.sop.md

Large diffs are not rendered by default.

.github/scripts/javascript/process-input.cjs

            
                      Original file line number
                      Diff line number
                      Diff line change
                  
    @@ -8,9 +8,10 @@ async function getIssueInfo(github, context, inputs) {
  
      const issueId = context.eventName === 'workflow_dispatch' 

        ? inputs.issue_id

        : context.payload.issue.number.toString();

      const commentBody = context.payload.comment?.body || '';

      const command = context.eventName === 'workflow_dispatch'

        ? inputs.command

        : (context.payload.comment.body.match(/^\/strands\s*(.*?)$/m)?.[1]?.trim() || '');

        : (commentBody.startsWith('/strands') ? commentBody.slice('/strands'.length).trim() : '');

      console.log(`Event: ${context.eventName}, Issue ID: ${issueId}, Command: "${command}"`);

    @@ -76,10 +77,25 @@ function buildPrompts(mode, issueId, isPullRequest, command, branchName, inputs)
  
      const scriptFile = scriptFiles[mode] || scriptFiles['refiner'];

      const systemPrompt = fs.readFileSync(scriptFile, 'utf8');

      // Extract the user's feedback/instructions after the mode keyword

      // e.g., "release-notes Move #123 to Major Features" -> "Move #123 to Major Features"

      const modeKeywords = {

        'release-notes': /^(?:release-notes|release notes)\s*/i,

        'implementer': /^implement\s*/i,

        'refiner': /^refine\s*/i

      };

      const modePattern = modeKeywords[mode];

      const userFeedback = modePattern ? command.replace(modePattern, '').trim() : command.trim();

      let prompt = (isPullRequest) 

        ? 'The pull request id is:'

        : 'The issue id is:';

      prompt += `${issueId}\n${command}\nreview and continue`;

      prompt += `${issueId}\n`;

      // If there's any user feedback beyond the command keyword, include it as the main instruction, 

      // otherwise default to "review and continue"

      prompt += userFeedback || 'review and continue';

      return { sessionId, systemPrompt, prompt };

    }

.github/scripts/python/agent_runner.py

            
                      Original file line number
                      Diff line number
                      Diff line change
                  
    @@ -39,7 +39,7 @@
  
    from str_replace_based_edit_tool import str_replace_based_edit_tool

    # Strands configuration constants

    STRANDS_MODEL_ID = "global.anthropic.claude-sonnet-4-5-20250929-v1:0"

    STRANDS_MODEL_ID = "global.anthropic.claude-opus-4-5-20251101-v1:0"

    STRANDS_MAX_TOKENS = 64000

    STRANDS_BUDGET_TOKENS = 8000

    STRANDS_REGION = "us-west-2"

    @@ -142,13 +142,12 @@ def run_agent(query: str):
  
    def main() -> None:

        """Main entry point for the agent runner."""

        try:

            # Read task from command line arguments

            if len(sys.argv) < 2:

                raise ValueError("Task argument is required")

            task = " ".join(sys.argv[1:])

            if not task.strip():

                raise ValueError("Task cannot be empty")

            # Prefer INPUT_TASK env var (avoids shell escaping issues), fall back to CLI args

            task = os.getenv("INPUT_TASK", "").strip()

            if not task and len(sys.argv) > 1:

                task = " ".join(sys.argv[1:]).strip()

            if not task:

                raise ValueError("Task is required (via INPUT_TASK env var or CLI argument)")

            print(f"🤖 Running agent with task: {task}")

            run_agent(task)

.github/workflows/LAMDBA_LAYERS_SOP.md

-Original file line number
+Diff line change
@@ -0,0 +1,31 @@
+    # Lambda Layers Standard Operating Procedures (SOP)
+    ## Overview
+    This document defines the standard operating procedures for managing Strands Agents Lambda layers across all AWS regions, Python versions, and architectures.
+    **Total: 136 individual Lambda layers** (17 regions × 2 architectures × 4 Python versions). All variants must maintain the same layer version number for each PyPI package version, with only one row per PyPI version appearing in documentation.
+    ## Deployment Process
+    ### 1. Initial Deployment
+. Run workflow with ALL options selected (default)
+. Specify PyPI package version
+. Type "Create Lambda Layer {package_version}" to confirm
+. All 136 individual layers deploy in parallel (4 Python × 2 arch × 17 regions)
+. Each layer gets its own unique name: `strands-agents-py{PYTHON_VERSION}-{ARCH}`
+    ### 2. Version Buffering for New Variants
+    When adding new variants (new Python version, architecture, or region):
+. **Determine target layer version**: Check existing variants to find the highest layer version
+. **Buffer deployment**: Deploy new variants multiple times until layer version matches existing variants
+. **Example**: If existing variants are at layer version 5, deploy new variant 5 times to reach version 5
+    ### 3. Handling Transient Failures
+    When some regions fail during deployment:
+. **Identify failed regions**: Check which combinations didn't complete successfully
+. **Targeted redeployment**: Use specific region/arch/Python inputs to redeploy failed combinations
+. **Version alignment**: Continue deploying until all variants reach the same layer version
+. **Verification**: Confirm all combinations have identical layer versions before updating docs

.github/workflows/publish-lambda-layer.yml

-Original file line number
+Diff line change
@@ -0,0 +1,167 @@
+    name: Publish PyPI Package to Lambda Layer
+    on:
+      workflow_dispatch:
+        inputs:
+          package_version:
+            description: 'Package version to download'
+            required: true
+            type: string
+          layer_version:
+            description: 'Layer version'
+            required: true
+            type: string
+          python_version:
+            description: 'Python version'
+            required: true
+            default: 'ALL'
+            type: choice
+            options: ['ALL', '3.10', '3.11', '3.12', '3.13']
+          architecture:
+            description: 'Architecture'
+            required: true
+            default: 'ALL'
+            type: choice
+            options: ['ALL', 'x86_64', 'aarch64']
+          region:
+            description: 'AWS region'
+            required: true
+            default: 'ALL'
+            type: choice
+            # Only non opt-in regions included for now
+            options: ['ALL', 'us-east-1', 'us-east-2', 'us-west-1', 'us-west-2', 'ap-south-1', 'ap-northeast-1', 'ap-northeast-2', 'ap-northeast-3', 'ap-southeast-1', 'ap-southeast-2', 'ca-central-1', 'eu-central-1', 'eu-west-1', 'eu-west-2', 'eu-west-3', 'eu-north-1', 'sa-east-1']
+          confirm:
+            description: 'Type "Create Lambda Layer {PyPI version}-layer{layer version}" to confirm publishing the layer'
+            required: true
+            type: string
+    jobs:
+      validate:
+        runs-on: ubuntu-latest
+        steps:
+          - name: Validate confirmation
+            run: |
+              CONFIRM="${{ inputs.confirm }}"
+              EXPECTED="Create Lambda Layer ${{ inputs.package_version }}-layer${{ inputs.layer_version }}"
+              if [ "$CONFIRM" != "$EXPECTED" ]; then
+                echo "Confirmation failed. You must type exactly '$EXPECTED' to proceed."
+                exit 1
+              fi
+              echo "Confirmation validated"
+      package-and-upload:
+        needs: validate
+        runs-on: ubuntu-latest
+        strategy:
+          matrix:
+            python-version: ${{ inputs.python_version == 'ALL' && fromJson('["3.10", "3.11", "3.12", "3.13"]') || fromJson(format('["{0}"]', inputs.python_version)) }}
+            architecture: ${{ inputs.architecture == 'ALL' && fromJson('["x86_64", "aarch64"]') || fromJson(format('["{0}"]', inputs.architecture)) }}
+            region: ${{ inputs.region == 'ALL' && fromJson('["us-east-1", "us-east-2", "us-west-1", "us-west-2", "ap-south-1", "ap-northeast-1", "ap-northeast-2", "ap-northeast-3", "ap-southeast-1", "ap-southeast-2", "ca-central-1", "eu-central-1", "eu-west-1", "eu-west-2", "eu-west-3", "eu-north-1", "sa-east-1"]') || fromJson(format('["{0}"]', inputs.region)) }}
+        permissions:
+          id-token: write
+        steps:
+          - name: Set up Python
+            uses: actions/setup-python@v4
+            with:
+              python-version: ${{ matrix.python-version }}
+          - name: Configure AWS credentials
+            uses: aws-actions/configure-aws-credentials@v4
+            with:
+              role-to-assume: ${{ secrets.STRANDS_LAMBDA_LAYER_PUBLISHER_ROLE }}
+              aws-region: ${{ matrix.region }}
+          - name: Create layer directory structure
+            run: |
+              mkdir -p layer/python
+          - name: Download and install package
+            run: |
+              pip install strands-agents==${{ inputs.package_version }} \
+                --python-version ${{ matrix.python-version }} \
+                --platform manylinux2014_${{ matrix.architecture }} \
+                -t layer/python/ \
+                --only-binary=:all:
+          - name: Create layer zip
+            run: |
+              cd layer
+              zip -r ../lambda-layer.zip .
+          - name: Upload to S3
+            run: |
+              PYTHON_VERSION="${{ matrix.python-version }}"
+              ARCH="${{ matrix.architecture }}"
+              REGION="${{ matrix.region }}"
+              LAYER_NAME="strands-agents-py${PYTHON_VERSION//./_}-${ARCH}"
+              ACCOUNT_ID=$(aws sts get-caller-identity --query Account --output text)
+              BUCKET_NAME="strands-layer-${ACCOUNT_ID}-${{ secrets.STRANDS_LAMBDA_LAYER_BUCKET_SALT }}-${REGION}"
+              LAYER_KEY="$LAYER_NAME/${{ inputs.package_version }}/layer${{ inputs.layer_version }}/lambda-layer.zip"
+              aws s3 cp lambda-layer.zip "s3://$BUCKET_NAME/$LAYER_KEY" --region "$REGION"
+              echo "Uploaded layer to s3://$BUCKET_NAME/$LAYER_KEY"
+      publish-layer:
+        needs: package-and-upload
+        runs-on: ubuntu-latest
+        strategy:
+          matrix:
+            python-version: ${{ inputs.python_version == 'ALL' && fromJson('["3.10", "3.11", "3.12", "3.13"]') || fromJson(format('["{0}"]', inputs.python_version)) }}
+            architecture: ${{ inputs.architecture == 'ALL' && fromJson('["x86_64", "aarch64"]') || fromJson(format('["{0}"]', inputs.architecture)) }}
+            region: ${{ inputs.region == 'ALL' && fromJson('["us-east-1", "us-east-2", "us-west-1", "us-west-2", "ap-south-1", "ap-northeast-1", "ap-northeast-2", "ap-northeast-3", "ap-southeast-1", "ap-southeast-2", "ca-central-1", "eu-central-1", "eu-west-1", "eu-west-2", "eu-west-3", "eu-north-1", "sa-east-1"]') || fromJson(format('["{0}"]', inputs.region)) }}
+        permissions:
+          id-token: write
+        steps:
+          - name: Configure AWS credentials
+            uses: aws-actions/configure-aws-credentials@v4
+            with:
+              role-to-assume: ${{ secrets.STRANDS_LAMBDA_LAYER_PUBLISHER_ROLE }}
+              aws-region: ${{ matrix.region }}
+          - name: Publish layer
+            run: |
+              PYTHON_VERSION="${{ matrix.python-version }}"
+              ARCH="${{ matrix.architecture }}"
+              REGION="${{ matrix.region }}"
+              LAYER_NAME="strands-agents-py${PYTHON_VERSION//./_}-${ARCH}"
+              ACCOUNT_ID=$(aws sts get-caller-identity --query Account --output text)
+              REGION_BUCKET="strands-layer-${ACCOUNT_ID}-${{ secrets.STRANDS_LAMBDA_LAYER_BUCKET_SALT }}-${REGION}"
+              LAYER_KEY="$LAYER_NAME/${{ inputs.package_version }}/layer${{ inputs.layer_version }}/lambda-layer.zip"
+              DESCRIPTION="PyPI package: strands-agents v${{ inputs.package_version }} (Python $PYTHON_VERSION, $ARCH)"
+              # Set compatible architecture based on matrix architecture
+              if [ "$ARCH" = "x86_64" ]; then
+                COMPATIBLE_ARCH="x86_64"
+              else
+                COMPATIBLE_ARCH="arm64"
+              fi
+              LAYER_OUTPUT=$(aws lambda publish-layer-version \
+                --layer-name $LAYER_NAME \
+                --description "$DESCRIPTION" \
+                --content S3Bucket=$REGION_BUCKET,S3Key=$LAYER_KEY \
+                --compatible-runtimes python${{ matrix.python-version }} \
+                --compatible-architectures $COMPATIBLE_ARCH \
+                --region "$REGION" \
+                --license-info Apache-2.0 \
+                --output json)
+              LAYER_ARN=$(echo "$LAYER_OUTPUT" | jq -r '.LayerArn')
+              LAYER_VERSION=$(echo "$LAYER_OUTPUT" | jq -r '.Version')
+              echo "Published layer version $LAYER_VERSION with ARN: $LAYER_ARN in region $REGION"
+              aws lambda add-layer-version-permission \
+                --layer-name $LAYER_NAME \
+                --version-number $LAYER_VERSION \
+                --statement-id public \
+                --action lambda:GetLayerVersion \
+                --principal '*' \
+                --region "$REGION"
+              echo "Successfully published layer version $LAYER_VERSION in region $REGION"

.gitignore

-Original file line number
+Diff line change
@@ Expand Up / @@ -14,3 +14,4 @@ repl_state @@
     .kiro
     uv.lock
     .audio_cache
+    CLAUDE.md

README.md

            
                      Original file line number
                      Diff line number
                      Diff line change
                  
    @@ -204,9 +204,9 @@ It's also available on GitHub via [strands-agents/tools](https://github.com/stra
  
    Build real-time voice and audio conversations with persistent streaming connections. Unlike traditional request-response patterns, bidirectional streaming maintains long-running conversations where users can interrupt, provide continuous input, and receive real-time audio responses. Get started with your first BidiAgent by following the [Quickstart](https://strandsagents.com/latest/documentation/docs/user-guide/concepts/experimental/bidirectional-streaming/quickstart) guide. 

    **Supported Model Providers:**

    - Amazon Nova Sonic (`amazon.nova-sonic-v1:0`)

    - Google Gemini Live (`gemini-2.5-flash-native-audio-preview-09-2025`)

    - OpenAI Realtime API (`gpt-realtime`)

    - Amazon Nova Sonic (v1, v2)

    - Google Gemini Live

    - OpenAI Realtime API

    **Quick Example:**

    @@ -219,7 +219,7 @@ from strands.experimental.bidi.tools import stop_conversation
  
    from strands_tools import calculator

    async def main():

        # Create bidirectional agent with audio model

        # Create bidirectional agent with Nova Sonic v2

        model = BidiNovaSonicModel()

        agent = BidiAgent(model=model, tools=[calculator, stop_conversation])

    @@ -241,14 +241,19 @@ if __name__ == "__main__":
  
    **Configuration Options:**

    ```python

    # Configure audio settings

    from strands.experimental.bidi.models import BidiNovaSonicModel

    # Configure audio settings and turn detection (v2 only)

    model = BidiNovaSonicModel(

        provider_config={

            "audio": {

                "input_rate": 16000,

                "output_rate": 16000,

                "voice": "matthew"

            },

            "turn_detection": {

                "endpointingSensitivity": "MEDIUM"  # HIGH, MEDIUM, or LOW

            },

            "inference": {

                "max_tokens": 2048,

                "temperature": 0.7

    @@ -263,6 +268,19 @@ audio_io = BidiAudioIO(
  
        input_buffer_size=10,

        output_buffer_size=10

    )

    # Text input mode (type messages instead of speaking)

    text_io = BidiTextIO()

    await agent.run(

        inputs=[text_io.input()],  # Use text input

        outputs=[audio_io.output(), text_io.output()]

    )

    # Multi-modal: Both audio and text input

    await agent.run(

        inputs=[audio_io.input(), text_io.input()],  # Speak OR type

        outputs=[audio_io.output(), text_io.output()]

    )

    ```

    ## Documentation

SECURITY.md

-Original file line number
+Diff line change
@@ -0,0 +1,20 @@
+    # Security Policy
+    ## Supported Versions
+    | Version | Supported          |
+    | ------- | ------------------ |
+    | 1.x.x   | :white_check_mark: |
+    | < 1.0   | :x:                |
+    ## Reporting Security Issues
+    Amazon Web Services (AWS) is dedicated to the responsible disclosure of security vulnerabilities.
+    We kindly ask that you **do not** open a public GitHub issue to report security concerns.
+    Instead, please submit the issue to the AWS Vulnerability Disclosure Program via [HackerOne](https://hackerone.com/aws_vdp) or send your report via [email](mailto:aws-security@amazon.com).
+    For more details, visit the [AWS Vulnerability Reporting Page](http://aws.amazon.com/security/vulnerability-reporting/).
+    Thank you in advance for collaborating with us to help protect our customers.

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

sdk main #7

Uh oh!

Diff view

Diff view

There are no files selected for viewing

Uh oh!

Uh oh!

Uh oh!

sdk main #7

Are you sure you want to change the base?

Uh oh!

sdk main #7

Uh oh!

Uh oh!

Diff view

Diff view

There are no files selected for viewing

Uh oh!

Uh oh!

Uh oh!