feat: Interactive service auditing in batches

[blairhyy-amazon]

Fixes

Summary

Changes

This PR implements interactive batch processing for service auditing in the CloudWatch Application Signals MCP server. The changes introduce a new workflow that automatically processes large service lists in manageable batches, allowing users to interactively decide whether to investigate findings or continue processing.

Key additions:

• New batch processing utilities (batch_processing_utils.py) - Core logic for creating, managing, and processing audit batches
• New batch tools (batch_tools.py) - MCP tools for continuing batch processing
• Enhanced audit_services function - Integrated batch processing for large service lists (>5 services)
• Interactive workflow - When findings are discovered, users can choose to investigate specific issues or continue processing
• Memory management - Session cleanup functionality to free resources
• Comprehensive test coverage - Full test suites for both batch processing utilities and tools

User experience

Before this change:

• Users auditing large numbers of services (>5) would receive overwhelming results all at once and wait for a long time before getting results
• No way to process services incrementally or interactively

After this change:

• Batch processing: Large service lists are processed in batches of 5
• Interactive decision making: When findings are discovered in a batch, users are presented with options to either investigate specific findings or continue processing the next batch
• Flexible workflow: Users maintain control over the audit process while benefiting from automated batching

Checklist

If your change doesn't seem to apply, please leave them unchecked.

• I have reviewed the contributing guidelines
• I have performed a self-review of this change
• Changes have been tested
• Changes are documented

Is this a breaking change? (Y/N)

RFC issue number:

Checklist:

• Migration process documented
• Implement warnings (if it can live side by side)

Acknowledgment

By submitting this pull request, I confirm that you can use, modify, copy, and redistribute this contribution, under the terms of the project license.

[codecov]

Codecov Report

❌ Patch coverage is 98.33333% with 2 lines in your changes missing coverage. Please review.
✅ Project coverage is 89.51%. Comparing base (620db3f) to head (0fdd7a0).

Files with missing lines	Patch %	Lines
...awslabs/cloudwatch_appsignals_mcp_server/server.py	89.47%	1 Missing and 1 partial ⚠️

Additional details and impacted files

@@            Coverage Diff             @@
##             main    #1530      +/-   ##
==========================================
+ Coverage   89.46%   89.51%   +0.05%     
==========================================
  Files         724      726       +2     
  Lines       50966    51082     +116     
  Branches     8145     8162      +17     
==========================================
+ Hits        45596    45727     +131     
+ Misses       3459     3446      -13     
+ Partials     1911     1909       -2     

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

🚀 New features to boost your workflow:

• ❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.
• 📦 JS Bundle Analysis: Save yourself from yourself by tracking and limiting bundle sizes in JS merges.

[yiyuan-he]

LGTM

[mxiamxia]

it looks the max sess limit is set to 1 by default, does it mean we don't support MCP to work with more than 1 session at the same time?

[mxiamxia]

is there a race condition risk here?

[mxiamxia]

Ideally, MCP server should be stateless. And it looks like you're generating random session id as key for the cache for each call. Does cache really work?