[notgitika]

Description

Generated Python Strands agents currently discard retrieved records below 0.5, including relevant records returned for a basic recall query in #2706. Set the generated relevance threshold to 0.0 so retrieval uses top_k without default score filtering. This covers the Python Strands, container, A2A, AG-UI, harness-export, and imported Bedrock Agent paths.

Use SEMANTIC and EPISODIC as the default strategies for memory-enabled runtime templates. Remove preference and summary retrieval from the default Python and TypeScript runtime templates. Harness exports continue to honor the strategies of explicitly configured existing memories and any supplied retrieval thresholds. Managed harness scaffolds keep their original empty managedMemoryConfiguration: {}.

These defaults apply to newly scaffolded runtime projects. Existing runtime projects can adopt them by updating their memory strategies and generated retrieval configuration.

Related Issue

Closes #2706

Documentation PR

Documentation is included here: the generated runtime template READMEs.

Type of Change

• Bug fix
• New feature
• Breaking change
• Documentation update
• Other (please describe):

Testing

• Full unit suite: 4,705 passed, 6 skipped, 0 failed across 311 files, run in three independent shards.

• Regression coverage verifies the provisioned strategies, episodic reflection namespace, matching Python/TypeScript retrieval namespaces, zero default thresholds, imported-agent code, and preservation of explicit 0 and 0.7 export thresholds.

• Built-CLI smoke tests verify that a managed harness emits managedMemoryConfiguration: {} and that Python Strands retains semantic/episodic strategies with zero relevance thresholds. The generated runtime templates are also covered by the unit regressions.

• bun run audit reports a pre-existing moderate minimatch finding, GHSA-g3hv-75wg-m249, through @aws/agent-inspector. The same finding reproduces with the upstream dependency files. This PR does not change package.json or bun.lock.

• Live AWS end-to-end tests were not run: the change concerns scaffolding defaults, which were verified through generated artifacts and the built CLI. No AWS resources were deployed.

• I ran bun test

• I ran the relevant end-to-end tests with bun run test:e2e, or explained why they are not applicable

• I ran bun run typecheck

• I ran bun run lint:check

• I ran bun run format:check

• I ran bun run build

• If I modified src/assets/, I updated affected snapshots with bun test <test-file> --update-snapshots and committed them

Checklist

• I have read the CONTRIBUTING document
• I have added any necessary tests that prove my fix is effective or my feature works
• I have updated the documentation accordingly
• I have added an appropriate example to the documentation to outline the feature, or no new docs are needed
• My changes generate no new warnings
• Any dependent changes have been merged and published

---

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

[github-actions]

Package Tarball

aws-agentcore-1.0.1.tgz

How to install

gh release download pr-2746-tarball --repo aws/agentcore-cli --pattern "*.tgz" --dir /tmp/pr-tarball
npm install -g /tmp/pr-tarball/aws-agentcore-1.0.1.tgz

[agentcore-devx-automation]

Claude Security Review: no high-confidence findings. (run)

[agentcore-devx-automation]

Claude Security Review: no high-confidence findings. (run)

[nborges-aws]

LGTM