Example for using Generative AI Toolkit with AWS AgentCore#68
Open
steditt wants to merge 36 commits into
Open
Conversation
- Add weather agent with MCP tools (forecast, alerts, current weather) - Add MCP server with FastMCP and weather API integration - Add CDK infrastructure for AgentCore Runtime deployment - Add minimal agent container example with Generative AI Toolkit - Add comprehensive test suite for all deployment scenarios - Include Docker Compose for local development and testing - Support local, containerized, and AgentCore Runtime deployments
- Add OAuth authentication with McpAuthHelper and ConfigLoader - Implement integration tests for auth_helper and config_loader - Add unified test configuration with CDK stack name from .env
…CP session handling - Replace CDK output fixtures with environment variables in conftest.py - Add MCP tool manager with proper event loop handling to fix session hangs - Remove minimal-agent-container and config_loader (deprecated) - Add comprehensive test suites for agent examples and local integration - Update pytest configuration to register custom markers - Streamline test architecture for better performance and maintainability
…mprehensive testing - Add Pydantic models (WeatherRequest, WeatherForecastRequest) with validation - Refactor MCP server to use FastMCP with automatic JSON schema generation - Add comprehensive test suite using SimpleMcpClient for schema validation - Update Dockerfile and requirements for proper deployment - Remove outdated documentation and cleanup test files
merge latest updates from upstream project
- Implement JWT bearer token authentication for agent invocations - Add Cognito User Pool integration with client credentials - Update tests to work with new JWT authentication requirements
- Replace agent user credentials with JWT token passthrough from client to MCP server - Remove AgentUser construct and associated Secrets Manager resources - Update agent.py to extract JWT from request headers and pass to MCP manager - Simplify simple_mcp_client.py to use JWT tokens directly - Remove simple_auth.py module and related OAuth environment variables - Clean up CDK infrastructure, IAM permissions, and Nag suppressions - Update tests and remove agent user credential handling This eliminates the need for dedicated service credentials and uses the actual user's JWT token throughout the request chain, simplifying architecture and improving security.
- Remove global MCP tool registration at module load time - Add lazy registration when JWT tokens become available in invoke() - Reset tools_registered flag when JWT token changes in McpToolManager - Eliminate confusing startup error messages about missing JWT tokens - Tools now register on-demand during first authenticated request
steditt
changed the title
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
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
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.
Description of changes:
Add example for using GAITK with AWS AgentCore.
By submitting this pull request, I confirm that you can use, modify, copy, and redistribute this contribution, under the terms of your choice.