To give a voice agent and a chat agent access to the same long-term user context, connect both Strands agents to one Amazon Bedrock AgentCore Memory resource. Derive the same stable actor ID for a signed-in person in both channels, use a separate session ID for each conversation, configure long-term memory strategies, and set retrieval rules for the records each agent should see. The shared resource is a documented capability; joining it to Amplify Gen 2 chat and a voice transport is a composition of documented patterns, not a single AWS end-to-end recipe.
What is shared—and what is not
AgentCore Memory organizes interaction data using a memory ID, an actor ID, and a session ID. For this design, the memory ID identifies the shared resource, the actor ID groups a user’s memories across channels, and the session ID identifies a conversation. AWS’s Strands integration documentation shows these as separate configuration values and describes memory use by agents. Sharing the resource alone does not establish who a user is, authorize access to that user’s records, or decide which records should be retrieved. Those are application responsibilities. See Get started with AgentCore Memory and the Strands memory integration.
In practical terms, the voice and chat frontends can remain separate. They use different connection and streaming paths, while their backend agents use the same memory resource and verified actor identity. AgentCore’s overview describes the service, and AWS separately documents chat and voice approaches discussed below.
Configure memory before connecting the agents
AgentCore can retain interaction events as short-term memory, but durable long-term records are generated only when memory strategies are configured. AWS documents semantic, summary, and user-preference strategies; choose the categories that match what the product should remember rather than assuming that event storage alone creates useful long-term recall. See AgentCore Memory strategies and Memory types.
Do these 3 things before closing this tab:
1Fix the driver behind crashes, sound loss and screen glitches2Repair Windows errors before they cause bigger problems3Scan for outdated or missing drivers - takes under a minute#1 Best Overall
- Your favorite music and content – Play music, audiobooks, and podcasts from Amazon Music, Apple Music, Spotify and others or via Bluetooth throughout your home.
- Alexa is happy to help – Ask Alexa for weather updates and to set hands-free timers, get answers to your questions and even hear jokes. Need a few extra minutes in the morning? Just tap your Echo Dot to snooze your alarm.
- Keep your home comfortable – Control compatible smart home devices with your voice and routines triggered by built-in motion or indoor temperature sensors. Create routines to automatically turn on lights when you walk into a room, or start a fan if the inside temperature goes above your comfort zone.
- Do more with device pairing – Fill your home with music using compatible Echo devices in different rooms, or create a home theatre system with Fire TV.
- Say goodbye to drop-offs and buffering - With eero Built-in, Echo Dot doubles as a mesh wifi extender, adding up to 1,000 sq. ft. of wifi coverage to your existing eero network.
- Create the AgentCore Memory resource. Configure one resource for the user context that should be shared by both channels.
- Configure one or more long-term strategies. AWS’s getting-started guide demonstrates a semantic strategy; its Strands integration example shows semantic, summary, and user-preference strategies together. Select strategies based on the durable facts, preferences, or summaries the application needs.
- Wait until the resource is active. Do not send events until the resource reaches the active status described in the getting-started guide.
- Set retrieval namespaces and rules. AWS’s Strands example uses
/summaries/{actorId}/{sessionId}/for session-specific summaries, and/preferences/{actorId}/and/facts/{actorId}/for user-level information. Treat these as an illustration, not a mandatory namespace scheme.
Retrieval scope affects what the model can use on a turn. Configure each agent to retrieve relevant namespaces and apply the per-namespace retrieval settings described in the Strands integration guide. Broad retrieval may pass more personal history than a particular turn needs; overly narrow retrieval may omit useful context.
Make identity stable and sessions deliberate
At a trusted backend boundary, validate the Cognito token or the token from your chosen identity provider, then derive the actor ID from the verified identity. Pass that actor ID to both channel agents. Mapping a Cognito subject to an actor ID is a sound implementation choice, not a literal mapping mandated by AWS. Never treat an actor ID supplied directly by the browser or voice client as proof of identity. The Amplify guide demonstrates Cognito-authenticated agent requests, while AgentCore’s memory setup and Strands configuration expose actor and session as distinct values.
Give each conversation its own session ID when the product should preserve channel-specific continuity. If a user explicitly resumes a particular conversation across channels, the application can instead use the same session identifier for that resumed interaction. A channel prefix can help with tracing, but session IDs should not contain sensitive information. Choose the session policy in the product layer; it is not implied by using one memory resource.
Rank #2
- Your favorite music and content – Play music, audiobooks, and podcasts from Amazon Music, Apple Music, Spotify and others or via Bluetooth throughout your home.
- Alexa is happy to help – Ask Alexa for weather updates and to set hands-free timers, get answers to your questions and even hear jokes. Need a few extra minutes in the morning? Just tap your Echo Dot to snooze your alarm.
- Keep your home comfortable – Control compatible smart home devices with your voice and routines triggered by built-in motion or indoor temperature sensors. Create routines to automatically turn on lights when you walk into a room, or start a fan if the inside temperature goes above your comfort zone.
- Do more with device pairing – Fill your home with music using compatible Echo devices in different rooms, or create a home theatre system with Fire TV.
- Say goodbye to drop-offs and buffering - With eero Built-in, Echo Dot doubles as a mesh wifi extender, adding up to 1,000 sq. ft. of wifi coverage to your existing eero network.
Connect both Strands agents to the shared resource
For each agent, configure AgentCoreMemoryConfig with the common memory ID and verified actor ID, the session ID selected for that conversation, and retrieval settings for the intended namespaces. Create an AgentCoreMemorySessionManager and pass it to the Strands Agent. The voice and chat configurations should share the resource and actor identity but can use different session IDs and retrieval scopes. The complete set of configuration fields and a working example are in the Strands SDK memory documentation.
Recommended Free Tools
This arrangement lets each agent draw on eligible user-level records and, where intended, session-specific summaries. It does not mean that every past message is automatically supplied to every turn: the configured strategy and retrieval behavior determine what long-term context is available.
Connect Amplify Gen 2 chat
The Amplify Gen 2 Agent Block documents an authenticated, streaming chat path using a Cognito user pool. It supports persisted conversations, and its example checks that a requested conversation belongs to the authenticated user before returning history. Keep that authorization check: a conversation identifier by itself does not authorize a history read.
Rank #3
- Meet Echo Dot Max: Experience rich room-filling sound that automatically adapts to your space and fine-tunes playback. Features a built-in smart home hub and Omnisense technology for highly personalized experiences.
- Music to your ears: With nearly 3x the bass versus Echo Dot (2022 release), it fits beautifully in any space, delivering your personal sound stage with deep bass and enhanced clarity. Listen to streaming services, such as Amazon Music, Apple Music, Spotify, and SiriusXM. Encore!
- Do more with device pairing: Connect compatible Echo smart speakers and smart displays in different rooms, or pair with a second Echo Dot Max to enjoy even richer sound. Pair your Echo Dot Max with compatible Fire TV devices to create a home theater system that brings scenes to life.
- Simple smart home control: Set routines, pair and control lights, locks, and thousands of smart home devices that work with Alexa without needing a separate smart home hub. With Omnisense technology, you can activate routines via temperature or presence detection.
- Say goodbye to drop-offs and buffering - With eero Built-in, Echo Dot Max doubles as a mesh wifi extender, adding up to 1,000 sq. ft. of wifi coverage to your existing eero network.
Use the chat backend to authenticate the request and establish the trusted actor identity before the Strands chat agent accesses AgentCore Memory. The Agent Block guide is not itself an end-to-end example of connecting its agent to shared AgentCore Memory; combining the Amplify chat pattern with the Strands memory integration is an implementation composition.
Connect the voice channel with its own transport
A browser or mobile voice client can connect to AgentCore Runtime using either bidirectional WebSocket streaming or WebRTC. AWS’s voice application guidance describes an Amplify-hosted frontend authenticating with Cognito and opening a SigV4-signed WebSocket to Runtime. Its architecture routes model tool calls through Lambda and AgentCore Gateway. Strands BidiAgent is another relevant option: AWS describes it as managing stream lifecycle, tool calls, and session handling in its voice-agent article.
Choose transport according to the client and media requirements, then route authenticated requests to the voice Strands agent configured for the shared memory resource. The transport does not replace the actor-identity and retrieval decisions used by the chat agent.
Rank #4
- Your favorite music and content – Play music, audiobooks, and podcasts from Amazon Music, Apple Music, Spotify and others or via Bluetooth throughout your home.
- Alexa is happy to help – Ask Alexa for weather updates and to set hands-free timers, get answers to your questions and even hear jokes. Need a few extra minutes in the morning? Just tap your Echo Dot to snooze your alarm.
- Keep your home comfortable – Control compatible smart home devices with your voice and routines triggered by built-in motion or indoor temperature sensors. Create routines to automatically turn on lights when you walk into a room, or start a fan if the inside temperature goes above your comfort zone.
- Do more with device pairing – Fill your home with music using compatible Echo devices in different rooms, or create a home theatre system with Fire TV.
- Say goodbye to drop-offs and buffering - With eero Built-in, Echo Dot doubles as a mesh wifi extender, adding up to 1,000 sq. ft. of wifi coverage to your existing eero network.
| Runtime transport | Documented fit | Considerations |
|---|---|---|
| WebSocket | Full-duplex text and audio streaming; AWS’s voice guidance uses a SigV4-signed WebSocket from an Amplify-hosted app. | Assess message protocol, signing, buffering, and interruption behavior for the application. |
| WebRTC | Real-time browser and mobile voice or video. | Assess UDP media and relay/network requirements, and confirm current Runtime and regional support. |
AWS announced WebRTC availability in fourteen AWS Regions on March 20, 2026. That figure describes the regions in that announcement, not performance, and availability can change. Check the current bidirectional streaming documentation and the March 2026 announcement for current regional details.
Independent reader supportYour contribution helps us test, update, and keep practical guides available for everyone.Account for asynchronous extraction and memory control
Long-term memory extraction is asynchronous after events are written. A voice turn may therefore be stored before its long-term insight is ready for retrieval. Keep the active turn context available to the current agent, and do not promise that a newly spoken detail will immediately appear in the next chat turn. This timing behavior is described in AgentCore’s memory types documentation.
Set product rules for which information is appropriate to retain, and consider ways for users to inspect or correct durable personal context. The documentation describes strategies and retrieval controls; it does not prescribe a complete user-control policy, so these choices belong to the product and its privacy requirements.
Free tools Windows power users keep installed
One-click scans. No signup required.
Best Value
- MEET ECHO SPOT - A sleek smart alarm clock with Alexa and big vibrant sound. Ready to help you wake up, wind down, and so much more.
- CUSTOMIZABLE SMART CLOCK - See time, weather, and song titles at a glance, control smart home devices, and more. Personalize your display with your favorite clock face and fun colors.
- BIG VIBRANT SOUND - Enjoy rich sound with clear vocals and deep bass. Just ask Alexa to play music, podcasts, and audiobooks. See song titles and touch to control your music.
- EASE INTO THE DAY - Set up an Alexa routine that gently wakes you with music and gradual light. Glance at the time, check reminders, or ask Alexa for weather updates.
- KEEP YOUR HOME COMFORTABLE - Control compatible smart home devices. Just ask Alexa to turn on lights or touch the screen to dim. Create routines that use motion detection to turn down the thermostat as you head out or open the blinds when you walk into a room.
Choose managed or self-managed strategies
AWS describes three broad approaches to strategies: built-in managed strategies, built-in overrides, and self-managed strategies. The trade-off is control versus operational work, not a choice between sharing and not sharing memory.
| Approach | What it offers | Trade-off |
|---|---|---|
| Built-in managed | Managed strategy behavior with limited customization. | AWS describes higher storage cost than the alternatives; no current price comparison is stated here. |
| Built-in overrides | Customization of prompts in the managed pipeline. | Retains the managed pipeline while allowing some control over behavior. |
| Self-managed | Control over extraction, consolidation, and schemas. | Requires the application team to provide and maintain infrastructure. |
Review the current strategy documentation before choosing; strategy behavior and service details may change.
Use long-term memory for personal context, RAG for current facts
Long-term memory is for personal or session context, such as preferences and prior interactions. It is not a substitute for retrieving current authoritative information from a maintained repository. AgentCore’s memory-versus-RAG comparison distinguishes these roles. For an answer that must reflect up-to-date policy, inventory, or other changing source material, use an appropriate retrieval source rather than relying on a remembered conversation.
Implementation checklist
- One AgentCore Memory resource is configured with the long-term strategies the experience needs.
- Both channel backends derive the same actor ID from a verified identity, rather than trusting a client-provided value.
- Session IDs reflect the intended conversation continuity; channel-specific sessions remain separate unless the user resumes a conversation.
- Each Strands agent uses
AgentCoreMemoryConfigand anAgentCoreMemorySessionManager, with appropriate retrieval scopes. - Amplify chat history reads verify conversation ownership.
- The voice frontend uses a supported authenticated Runtime transport, with region and implementation details checked against current AWS documentation.
- The experience handles asynchronous extraction and has product rules for retained and retrievable personal context.
AWS documentation and Amplify Blocks evolve. The cited pages were consulted on October 4, 2026; the Amplify Agent Block page reports an update date of July 2, 2026, and AWS’s voice article is dated May 19, 2026. Recheck current API names, permissions, supported models, package versions, and regional availability before implementation.
What’s actually slowing this PC down?
Pick the symptom - the matching free tool is one click away.
Quick Recap
Product prices and availability are accurate as of the date/time indicated and are subject to change. Any price and availability information displayed on Amazon at the time of purchase will apply.
