feat: add support for LTM metadata

[nborges-aws]

Description

Adds support for AgentCore Memory's indexed metadata keys to the CLI. Indexed keys are declared at the memory level and enable filtering LTM records on retrieval. Each key has a name and a value type (STRING, STRINGLIST, NUMBER). There is a 10 key max per memory.

Schema:

• New IndexedKeyTypeSchema enum and IndexedKeySchema in agentcore-project.ts
• MemorySchema gains optional indexedKeys field with 10 key max, enforced uniqueness, and a superRefine that rejects indexedKeys when no strategies are present.

Non-interactive CLI:

• Repeatable --indexed-key <key:TYPE> flag on agentcore add memory. Accumulates into a list, parsed by a shared parseIndexedKeyArg helper in indexed-key-parser.ts

Interactive TUI:

• New step in the agentcore add memory wizard that conditionally appears only when at least one LTM strategy is selected. Step has four sub-states: prompt → keyName → keyType → addAnother, with a tally of ✓ name (TYPE) shown above each input.
• Inline validation matches the schema. Auto-finishes the step at the 10-key cap. Esc/back navigation walks back through previous inputs.
• "Clear keys" option for convenience bulk removing key selections
• Selecting a non-LTM strategy mix on the strategies step clears any previously-collected indexed keys, so a stale list never surfaces
• "Indexed Keys" line on the confirm screen lists the saved keys.

Type of Change

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

Testing

How have you tested the change?

• I ran npm run test:unit and npm run test:integ
• I ran npm run typecheck
• I ran npm run lint
• If I modified src/assets/, I ran npm run test:update-snapshots and committed the updated snapshots

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: depends on https://github.com/aws/agentcore-l3-cdk-constructs/pull/224

---

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-0.14.0.tgz

How to install

npm install https://github.com/aws/agentcore-cli/releases/download/pr-1281-tarball/aws-agentcore-0.14.0.tgz

[github-actions]

Coverage Report

Status	Category	Percentage	Covered / Total
🔵	Lines	43.53%	9220 / 21179
🔵	Statements	42.78%	9780 / 22857
🔵	Functions	40.19%	1590 / 3956
🔵	Branches	40.31%	6003 / 14891

Generated in workflow #3051 for commit 9ae5e89 by the Vitest Coverage Report Action

[agentcore-cli-automation]

Nice feature with a tight scope and good unit-test coverage on the parser/validator. A few things to address before merging:

1. The TUI re-hydrates indexed-key state by calling setState during render — should be a useEffect to avoid Strict-Mode warnings and double-render surprises.
2. The TUI key-name validation duplicates the regex/length rules that already live in indexed-key-parser.ts. Reuse the shared constants so they can't drift.
3. add.memory telemetry doesn't capture anything about indexed keys. Per src/cli/telemetry/README.md new features should be instrumented (e.g. indexed_key_count / has_indexed_keys).

There's also a minor UX papercut on Esc-back from the key-type sub-step (the previously-typed key name is silently lost) — non-blocking but worth tightening up while you're in here.