Add comprehensive agentic memory API documentation for OpenSearch 3.3 #11169

dhrubo-os · 2025-10-01T23:38:28Z

Add long-term memory APIs: search, update, delete
Add memory history APIs: search, delete
Add general memory APIs: get, update, delete, search by type
Update add-memory API to include data type memory and structured data
Update index with complete API structure
Support for conversation memory, data memory, and trace data
Include namespace-based organization and tagging system
Remove experimental warning as agentic memory is GA in 3.3

Description

Describe what this change achieves.

Issues Resolved

Closes #11162

Version

3.3

Frontend features

If you're submitting documentation for an OpenSearch Dashboards feature, add a video that shows how a user will interact with the UI step by step. A voiceover is optional.

Checklist

By submitting this pull request, I confirm that my contribution is made under the terms of the Apache 2.0 license and subject to the Developers Certificate of Origin.
For more information on following Developer Certificate of Origin and signing off your commits, please check here.

github-actions · 2025-10-01T23:38:36Z

Thank you for submitting your PR. The PR states are In progress (or Draft) -> Tech review -> Doc review -> Editorial review -> Merged.

Before you submit your PR for doc review, make sure the content is technically accurate. If you need help finding a tech reviewer, tag a maintainer.

When you're ready for doc review, tag the assignee of this PR. The doc reviewer may push edits to the PR directly or leave comments and editorial suggestions for you to address (let us know in a comment if you have a preference). The doc reviewer will arrange for an editorial review.

kolchfa-aws · 2025-10-02T13:29:29Z

@dhrubo-os Thank you for the PR! When you connect it to an issue, you have to leave "Closes", "Fixes", or "Resolves" so it's properly linked (in this case, Closes #11162). I updated the PR description.

kolchfa-aws · 2025-10-09T18:23:18Z

Hey @dhrubo-os , let me know when this PR is ready for doc review. We have 3 days left before the release.

dhrubo-os · 2025-10-10T03:35:39Z

Hey @dhrubo-os , let me know when this PR is ready for doc review. We have 3 days left before the release.

Yes please start reviewing the PR. Thanks

ylwu-amzn · 2025-10-10T04:09:23Z

_ml-commons-plugin/api/agentic-memory-apis/delete-memory-by-type.md

+
+## Path parameters
+
+| Parameter | Data type | Description |


https://docs.opensearch.org/latest/ml-commons-plugin/api/agentic-memory-apis/create-memory-container/

Currently we have a separate column Required/Optional, can we follow the same format?

ylwu-amzn · 2025-10-10T04:13:25Z

_ml-commons-plugin/api/agentic-memory-apis/add-memory.md

-{: .warning}

-Use this API to add an agentic memory to a [memory container]({{site.url}}{{site.baseurl}}/ml-commons-plugin/api/agentic-memory-apis/create-memory-container). You can create a memory in one of the following modes (controlled by the `infer` parameter):
+Use this API to add an agentic memory to a [memory container]({{site.url}}{{site.baseurl}}/ml-commons-plugin/api/agentic-memory-apis/create-memory-container). You can create memories in two types:


Suggested change

Use this API to add an agentic memory to a [memory container]({{site.url}}{{site.baseurl}}/ml-commons-plugin/api/agentic-memory-apis/create-memory-container). You can create memories in two types:

Use this API to add an agentic memory to a [memory container]({{site.url}}{{site.baseurl}}/ml-commons-plugin/api/agentic-memory-apis/create-memory-container). You can provide memory payload in two types:

ylwu-amzn · 2025-10-10T04:14:08Z

_ml-commons-plugin/api/agentic-memory-apis/add-memory.md

-Use this API to add an agentic memory to a [memory container]({{site.url}}{{site.baseurl}}/ml-commons-plugin/api/agentic-memory-apis/create-memory-container). You can create a memory in one of the following modes (controlled by the `infer` parameter):
+Use this API to add an agentic memory to a [memory container]({{site.url}}{{site.baseurl}}/ml-commons-plugin/api/agentic-memory-apis/create-memory-container). You can create memories in two types:
+
+- **Conversation memory** -- Stores conversational messages between users and assistants. Can be processed (when `infer` is `true`) to extract facts or stored as raw messages.


Suggested change

- **Conversation memory** -- Stores conversational messages between users and assistants. Can be processed (when `infer` is `true`) to extract facts or stored as raw messages.

- **conversational** -- Stores conversational messages between users and assistants.

ylwu-amzn · 2025-10-10T04:14:57Z

_ml-commons-plugin/api/agentic-memory-apis/add-memory.md

+
+- **Conversation memory** -- Stores conversational messages between users and assistants. Can be processed (when `infer` is `true`) to extract facts or stored as raw messages.
+
+- **Data memory** -- Stores structured, non-conversational data such as agent state, checkpoints, or reference information.


Suggested change

- **Data memory** -- Stores structured, non-conversational data such as agent state, checkpoints, or reference information.

- **data** -- Stores extra messages, structured, non-conversational data such as agent state, checkpoints, or reference information.

ylwu-amzn · 2025-10-10T04:17:40Z

_ml-commons-plugin/api/agentic-memory-apis/add-memory.md

+
+Memory processing modes (controlled by the `infer` parameter):

 - Fact memory -- A processed representation of the message. The large language model (LLM) associated with the memory container extracts and stores key factual information or knowledge from the original text.


Suggested change

- Fact memory -- A processed representation of the message. The large language model (LLM) associated with the memory container extracts and stores key factual information or knowledge from the original text.

- **infer=true** -- Use large language model (LLM) to extract key information or knowledge from the messages.

ylwu-amzn · 2025-10-10T04:19:01Z

_ml-commons-plugin/api/agentic-memory-apis/add-memory.md


 - Fact memory -- A processed representation of the message. The large language model (LLM) associated with the memory container extracts and stores key factual information or knowledge from the original text.

 - Raw message memory -- The unprocessed message content.


Suggested change

- **infer=false** -- Only store raw messages and data in working memory.

ylwu-amzn · 2025-10-10T04:20:02Z

_ml-commons-plugin/api/agentic-memory-apis/add-memory.md

-`session_id` | String | Optional | The session ID associated with the memory.
-`agent_id` | String | Optional | The agent ID associated with the memory.
-`infer` | Boolean | Optional | Controls whether the LLM infers context from messages. Default is `true`. When `true`, the LLM extracts factual information from the original text and stores it as the memory. When `false`, the memory contains the unprocessed message and you must explicitly specify the `role` in each message. 
+`messages` | List | Conditional | A list of messages for conversation memory. Each message requires `content` and may include a `role` (commonly, `user` or `assistant`) when `infer` is set to `true`. Required for `memory_type` of `conversation`.


Suggested change

`messages` | List | Conditional | A list of messages for conversation memory. Each message requires `content` and may include a `role` (commonly, `user` or `assistant`) when `infer` is set to `true`. Required for `memory_type` of `conversation`.

`messages` | List | Conditional | A list of messages for conversational payload. Each message requires `content` and may include a `role` (commonly, `user` or `assistant`) when `infer` is set to `true`. Required for `payload_type` of `conversational`.

ylwu-amzn · 2025-10-10T04:21:39Z

_ml-commons-plugin/api/agentic-memory-apis/add-memory.md

-`agent_id` | String | Optional | The agent ID associated with the memory.
-`infer` | Boolean | Optional | Controls whether the LLM infers context from messages. Default is `true`. When `true`, the LLM extracts factual information from the original text and stores it as the memory. When `false`, the memory contains the unprocessed message and you must explicitly specify the `role` in each message. 
+`messages` | List | Conditional | A list of messages for conversation memory. Each message requires `content` and may include a `role` (commonly, `user` or `assistant`) when `infer` is set to `true`. Required for `memory_type` of `conversation`.
+`structured_data` | Object | Conditional | Structured data content for data memory. Required for `memory_type` of `data`.


Add a new row for binary_data

And metadata also?

ylwu-amzn · 2025-10-10T04:21:58Z

_ml-commons-plugin/api/agentic-memory-apis/add-memory.md

-`infer` | Boolean | Optional | Controls whether the LLM infers context from messages. Default is `true`. When `true`, the LLM extracts factual information from the original text and stores it as the memory. When `false`, the memory contains the unprocessed message and you must explicitly specify the `role` in each message. 
+`messages` | List | Conditional | A list of messages for conversation memory. Each message requires `content` and may include a `role` (commonly, `user` or `assistant`) when `infer` is set to `true`. Required for `memory_type` of `conversation`.
+`structured_data` | Object | Conditional | Structured data content for data memory. Required for `memory_type` of `data`.
+`memory_type` | String | Required | The type of memory: `conversation` or `data`.


ylwu-amzn · 2025-10-10T04:22:42Z

_ml-commons-plugin/api/agentic-memory-apis/add-memory.md

+`messages` | List | Conditional | A list of messages for conversation memory. Each message requires `content` and may include a `role` (commonly, `user` or `assistant`) when `infer` is set to `true`. Required for `memory_type` of `conversation`.
+`structured_data` | Object | Conditional | Structured data content for data memory. Required for `memory_type` of `data`.
+`memory_type` | String | Required | The type of memory: `conversation` or `data`.
+`namespace` | Object | Optional | Namespace context for organizing memories (e.g., `user_id`, `session_id`, `agent_id`).


Explain more details

If session_id not exists in namespace will create a new session and use the new session's id

ylwu-amzn · 2025-10-10T04:23:50Z

_ml-commons-plugin/api/agentic-memory-apis/add-memory.md

+`structured_data` | Object | Conditional | Structured data content for data memory. Required for `memory_type` of `data`.
+`memory_type` | String | Required | The type of memory: `conversation` or `data`.
+`namespace` | Object | Optional | Namespace context for organizing memories (e.g., `user_id`, `session_id`, `agent_id`).
+`metadata` | Object | Optional | Additional metadata for the memory (e.g., `status`, `branch`, custom fields).


From the code looks like it's Map<String, Object>

ylwu-amzn · 2025-10-10T04:24:21Z

_ml-commons-plugin/api/agentic-memory-apis/add-memory.md

+`memory_type` | String | Required | The type of memory: `conversation` or `data`.
+`namespace` | Object | Optional | Namespace context for organizing memories (e.g., `user_id`, `session_id`, `agent_id`).
+`metadata` | Object | Optional | Additional metadata for the memory (e.g., `status`, `branch`, custom fields).
+`session_id` | String | Optional | The session ID associated with the memory. Deprecated in favor of using `namespace.session_id`.


Remove these two lines: 45, 46

ylwu-amzn · 2025-10-10T04:32:04Z

_ml-commons-plugin/api/agentic-memory-apis/add-memory.md

+`metadata` | Object | Optional | Additional metadata for the memory (e.g., `status`, `branch`, custom fields).
+`session_id` | String | Optional | The session ID associated with the memory. Deprecated in favor of using `namespace.session_id`.
+`agent_id` | String | Optional | The agent ID associated with the memory. Deprecated in favor of using `namespace.agent_id`.
+`infer` | Boolean | Optional | Controls whether the LLM infers context from messages. Default is `true` for conversation memory, `false` for data memory. When `true`, the LLM extracts factual information from the original text and stores it as the memory. When `false`, the memory contains the unprocessed message and you must explicitly specify the `role` in each message. 


Suggested change

`infer` | Boolean | Optional | Controls whether the LLM infers context from messages. Default is `true` for conversation memory, `false` for data memory. When `true`, the LLM extracts factual information from the original text and stores it as the memory. When `false`, the memory contains the unprocessed message and you must explicitly specify the `role` in each message.

`infer` | Boolean | Optional | Controls whether use LLM to extract key information from messages. Default is `false`. When `true`, the LLM extracts key information from the original text and stores it as the memory.

ylwu-amzn · 2025-10-10T04:34:14Z

_ml-commons-plugin/api/agentic-memory-apis/add-memory.md

+`structured_data` | Object | Conditional | Structured data content for data memory. Required for `memory_type` of `data`.
+`memory_type` | String | Required | The type of memory: `conversation` or `data`.
+`namespace` | Object | Optional | Namespace context for organizing memories (e.g., `user_id`, `session_id`, `agent_id`).
+`metadata` | Object | Optional | Additional metadata for the memory (e.g., `status`, `branch`, custom fields).


Add a new row for tags

ylwu-amzn · 2025-10-10T04:34:55Z

_ml-commons-plugin/api/agentic-memory-apis/add-memory.md

-## Example request
+## Example requests
+
+### Conversation memory


Suggested change

### Conversation memory

### Conversational payload

ylwu-amzn · 2025-10-10T20:53:40Z

_ml-commons-plugin/api/agentic-memory-apis/create-memory-container.md

+`embedding_model_id` | String | Optional | The embedding model ID.
+`embedding_dimension` | Integer | Optional | The dimension of the embedding model. Required if `embedding_model_type` is `TEXT_EMBEDDING`.
+`llm_id` | String | Optional | The LLM model ID for processing and inference.
+`index_prefix` | String | Optional | Custom prefix for the memory indices. If not specified, a default prefix is used: if `use_system_index` is `true`, use `default` as index prefix; if `use_system_index` is `false`, use 8 bit random UUID as index prefix.


We also support such fields

disable_history, bool

disable_session, bool

max_infer_size: int ,

index_settings , Map<String, Map<String, Object>>
Example of index_settings

"index_settings": { "session_index" : { "index": { "number_of_shards": "2", "number_of_replicas": "2" } }, "short_term_memory_index" : { "index": { "number_of_shards": "2", "number_of_replicas": "2" } }, "long_term_memory_index" : { "index": { "number_of_shards": "2", "number_of_replicas": "2" } }, "long_term_memory_history_index" : { "index": { "number_of_shards": "2", "number_of_replicas": "2" } } }

ylwu-amzn · 2025-10-10T20:56:24Z

_ml-commons-plugin/api/agentic-memory-apis/create-memory-container.md

+
+Field | Data type | Required/Optional | Description
+:--- | :--- | :--- | :---
+`type` | String | Required | The strategy type: `SEMANTIC`, `USER_PREFERENCE`, or `SUMMARY`.


add enabled field, bool type, optional

- Add long-term memory APIs: search, update, delete - Add memory history APIs: search, delete - Add general memory APIs: get, update, delete, search by type - Update add-memory API to include data type memory and structured data - Update index with complete API structure - Support for conversation memory, data memory, and trace data - Include namespace-based organization and tagging system - Remove experimental warning as agentic memory is GA in 3.3 Signed-off-by: Dhrubo Saha <[email protected]>

- Remove redundant API files in favor of unified memory APIs - Delete legacy files: delete-long-term-memory.md, delete-memory-history.md, search-long-term-memory.md, search-memory-history.md, update-long-term-memory.md, search-memory.md, delete-memory.md, update-memory.md, get-memory.md - Enhance unified APIs with comprehensive examples for all memory types (sessions, working, long-term, history) - Add metadata field documentation to add-memory.md and get-memory-by-type.md - Update response examples and field documentation with actual API responses - Update index.md to reflect clean unified API structure - Fix memory type parameter names (sessions vs session) for consistency Signed-off-by: Dhrubo Saha <[email protected]>

- Add missing update-memory-container.md and search-memory-container.md files - Update create-memory-container.md with advanced features from Postman collection: * Multiple strategy support (SEMANTIC, USER_PREFERENCE, SUMMARY) * Strategy configuration with llm_result_path * Index configuration (index_prefix, use_system_index) * Remove experimental warning (GA in 3.3) - Remove experimental feature reference from index.md - All APIs now match actual implementation from Postman collection Signed-off-by: Dhrubo Saha <[email protected]>

- Add default llm_result_path value: $.output.message.content[0].text (Bedrock Converse format) - Document LLM connector requirement for system_prompt and user_prompt parameters - Reference ml-commons PR opensearch-project#4283 for standardized connector format - Clarify that default is optimized for Bedrock Converse API Signed-off-by: Dhrubo Saha <[email protected]>

- Update add-memory.md terminology: memory_type -> payload_type, conversational/data types - Add Required/Optional column format to parameter tables - Update version labels from 3.2 to 3.3 for new unified APIs - Add binary_data and tags fields with correct data types - Fix metadata field representation as string in response examples - Add detailed explanations for namespace session creation behavior - Update create-memory-container.md with system_prompt and llm_id strategy fields - Add detailed index_prefix default behavior explanation - Show multiple namespace field examples in strategy configuration Signed-off-by: Dhrubo Saha <[email protected]>

- Add plugins.ml_commons.agentic_memory_enabled setting to cluster-settings.md - Enables/disables agentic memory functionality for AI agents - Includes session, working, long-term memory, and memory history features Signed-off-by: Dhrubo Saha <[email protected]>

Signed-off-by: Dhrubo Saha <[email protected]>

b4sjoo · 2025-10-10T21:12:51Z

_ml-commons-plugin/api/agentic-memory-apis/update-memory-by-type.md

+
+| Field | Data type | Description |
+| :--- | :--- | :--- |
+| `additional_info` | Object | Additional metadata to associate with the session. Optional. |


The summary field is updatable

b4sjoo · 2025-10-10T21:14:51Z

_ml-commons-plugin/api/agentic-memory-apis/update-memory-by-type.md

+
+| Field | Data type | Description |
+| :--- | :--- | :--- |
+| `memory` | String | The updated memory content. Optional. |


Guide user that update memory field will cause embedding automatically updated?

b4sjoo · 2025-10-10T21:15:27Z

_ml-commons-plugin/api/agentic-memory-apis/update-memory-by-type.md

+| Field | Data type | Description |
+| :--- | :--- | :--- |
+| `messages` | Array | Updated conversation messages (for conversation type). Optional. |
+| `structured_data` | Object | Updated structured data content (for data type). Optional. |


metadata, binary_data also can be updated

b4sjoo · 2025-10-10T21:19:16Z

_ml-commons-plugin/api/agentic-memory-apis/update-memory-container.md

+
+| Field | Data type | Required/Optional | Description |
+| :--- | :--- | :--- | :--- |
+| `name` | String | Optional | The updated name of the memory container. |


llm_id, backend_roles, and [strategies, embedding_model_id, embedding_model_type, embedding_dimension] fields inside configuration object all can be updated

In the meantime, update strategies with id will only update that specific strategy, if we specified a strategy without id, a new strategy will be created

Adding new backend_roles in backend_roles will let grant new user read or write access with that backend_roles.

backend_roles and namespace field (inside strategies) are update by overwritten, so user need to make sure that original field is included if he wants to keep it

embedding_model_id, embedding_model_type, embedding_dimension fields are only allowed to update if the long-term memory index has not created for this memory container yet. That means once we have a long-term memory index with the index_prefix specified in this memory container created, embedding_model_id, embedding_model_type, embedding_dimension cannot be updated.

ylwu-amzn · 2025-10-10T22:54:17Z

_ml-commons-plugin/api/agentic-memory-apis/create-memory-container.md

+
+Field | Data type | Required/Optional | Description
+:--- | :--- | :--- | :---
+`type` | String | Required | The strategy type: `SEMANTIC`, `USER_PREFERENCE`, or `SUMMARY`.


Memory strategy also support enabled field

ylwu-amzn · 2025-10-10T23:01:25Z

_ml-commons-plugin/cluster-settings.md

+### Setting
+
+```yaml
+plugins.ml_commons.max_memories_per_container: 10000


remove this, we don't have this setting

ylwu-amzn · 2025-10-10T23:01:32Z

_ml-commons-plugin/cluster-settings.md

+### Setting
+
+```yaml
+plugins.ml_commons.max_memory_containers_per_user: 100


remove this, we don't have this setting

ylwu-amzn · 2025-10-10T23:03:50Z

_ml-commons-plugin/api/agentic-memory-apis/update-memory-container.md

+| Field | Data type | Required/Optional | Description |
+| :--- | :--- | :--- | :--- |
+| `name` | String | Optional | The updated name of the memory container. |
+| `description` | String | Optional | The updated description of the memory container. |


I think we support more fields in update container API. @b4sjoo can you help review?

ylwu-amzn · 2025-10-10T23:06:34Z

_ml-commons-plugin/api/agentic-memory-apis/update-memory-by-type.md

+
+The request fields vary depending on the memory type being updated:
+
+### For session memory


Suggest update the doc based on this code https://github.com/opensearch-project/ml-commons/blob/main/plugin/src/main/java/org/opensearch/ml/action/memorycontainer/memory/TransportUpdateMemoryAction.java#L144-L156

ylwu-amzn · 2025-10-10T23:07:19Z

_ml-commons-plugin/api/agentic-memory-apis/update-memory-by-type.md

+
+| Field | Data type | Description |
+| :--- | :--- | :--- |
+| `additional_info` | Object | Additional metadata to associate with the session. Optional. |


We support updating these fields for session

summary

metadata

agents

additional_info

ylwu-amzn · 2025-10-10T23:07:57Z

_ml-commons-plugin/api/agentic-memory-apis/update-memory-by-type.md

+| Field | Data type | Description |
+| :--- | :--- | :--- |
+| `messages` | Array | Updated conversation messages (for conversation type). Optional. |
+| `structured_data` | Object | Updated structured data content (for data type). Optional. |


Add new row for

binary_data

metadata

I also added these

ylwu-amzn · 2025-10-10T23:08:46Z

_ml-commons-plugin/api/agentic-memory-apis/update-memory-by-type.md

+| `messages` | Array | Updated conversation messages (for conversation type). Optional. |
+| `structured_data` | Object | Updated structured data content (for data type). Optional. |
+| `tags` | Object | Updated tags for categorization. Optional. |
+| `additional_info` | Object | Additional metadata. Optional. |


remove this line

Signed-off-by: Dhrubo Saha <[email protected]>

ylwu-amzn · 2025-10-11T07:53:32Z

_ml-commons-plugin/api/agentic-memory-apis/add-memory.md

+`messages` | List                 | Conditional | A list of messages for conversational payload. Each message requires `content`. Required for `payload_type` of `conversational`.
+`structured_data` | Map<String, Object>  | Conditional | Structured data content for data memory. Required for `payload_type` of `data`.
+`binary_data` | String               | Optional | Binary data content encoded as base64 string for binary payloads.
+`payload_type` | String               | Required | The type of payload: `conversational` or `data`.


This is optional, default is conversational

ylwu-amzn · 2025-10-11T07:55:54Z

_ml-commons-plugin/api/agentic-memory-apis/add-memory.md

+Field | Data type            | Required/Optional | Description
+:--- |:---------------------| :--- | :---
+`messages` | List                 | Conditional | A list of messages for conversational payload. Each message requires `content`. Required for `payload_type` of `conversational`.
+`structured_data` | Map<String, Object>  | Conditional | Structured data content for data memory. Required for `payload_type` of `data`.


This is optional, even for data type.

ylwu-amzn · 2025-10-11T07:59:58Z

_ml-commons-plugin/api/agentic-memory-apis/create-memory-container.md

-
 Use this API to create a memory container to hold agentic memories. The container can have two model types associated with it:

 - A text embedding model for vectorizing the message so it can be searched. Use a text embedding model for dense vector embeddings or a sparse encoding model for sparse vector formats. If no embedding model is specified, messages are stored but cannot be used for vector-based searches.


Long term memory requires both an LLM model and embedding model to be configured.

ylwu-amzn · 2025-10-11T08:03:13Z

_ml-commons-plugin/api/agentic-memory-apis/create-memory-container.md


 ### LLM 

 Currently, agentic memory supports only the Anthropic Claude model for LLM capabilities. Starting with OpenSearch 3.3, this feature will be generally available and will include support for additional LLM providers, such as OpenAI.


Remove this line? Or tune it, as we do support multiple LLMs in 3.3

ylwu-amzn · 2025-10-11T08:04:16Z

_ml-commons-plugin/api/agentic-memory-apis/create-memory-container.md

    "function_name": "remote",
    "description": "test model",
    "connector": {
        "name": "Amazon Bedrock Connector: embedding",


This name is not correct, not embedding

ylwu-amzn · 2025-10-11T08:05:07Z

_ml-commons-plugin/api/agentic-memory-apis/create-memory-container.md

                "content-type": "application/json"
            },
            "url": "https://bedrock-runtime.${parameters.region}.amazonaws.com/model/${parameters.model}/invoke",
            "request_body": """{ "system": "${parameters.system_prompt}", "anthropic_version": "${parameters.anthropic_version}", "max_tokens": ${parameters.max_tokens}, "temperature": ${parameters.temperature}, "messages": ${parameters.messages} }"""


Change the request body

ylwu-amzn · 2025-10-11T08:05:39Z

_ml-commons-plugin/api/agentic-memory-apis/create-memory-container.md

            "headers": {
                "content-type": "application/json"
            },
            "url": "https://bedrock-runtime.${parameters.region}.amazonaws.com/model/${parameters.model}/invoke",


Change to use converse API ? The default llm_result_path is for bedrock converse API

dhrubo-os requested review from AMoo-Miki, dlvenable, epugh, kolchfa-aws, natebower and sumobrian as code owners October 1, 2025 23:38

github-actions bot assigned kolchfa-aws Oct 1, 2025

dhrubo-os force-pushed the main branch from d45e504 to b08069d Compare October 1, 2025 23:39

kolchfa-aws added Tech review PR: Tech review in progress release-notes PR: Include this PR in the automated release notes v3.3.0 labels Oct 2, 2025

ylwu-amzn reviewed Oct 10, 2025

View reviewed changes

dhrubo-os added 7 commits October 10, 2025 14:02

update path parameters

f4d4bbf

Signed-off-by: Dhrubo Saha <[email protected]>

b4sjoo reviewed Oct 10, 2025

View reviewed changes

ylwu-amzn reviewed Oct 10, 2025

View reviewed changes

addressed comments

1340f34

Signed-off-by: Dhrubo Saha <[email protected]>

dhrubo-os force-pushed the main branch from cdcdd18 to 1340f34 Compare October 11, 2025 01:03

ylwu-amzn reviewed Oct 11, 2025

View reviewed changes

	Use this API to add an agentic memory to a [memory container]({{site.url}}{{site.baseurl}}/ml-commons-plugin/api/agentic-memory-apis/create-memory-container). You can create memories in two types:
	Use this API to add an agentic memory to a [memory container]({{site.url}}{{site.baseurl}}/ml-commons-plugin/api/agentic-memory-apis/create-memory-container). You can provide memory payload in two types:

	- Conversation memory -- Stores conversational messages between users and assistants. Can be processed (when `infer` is `true`) to extract facts or stored as raw messages.
	- conversational -- Stores conversational messages between users and assistants.


		- Conversation memory -- Stores conversational messages between users and assistants. Can be processed (when `infer` is `true`) to extract facts or stored as raw messages.

		- Data memory -- Stores structured, non-conversational data such as agent state, checkpoints, or reference information.

	- Data memory -- Stores structured, non-conversational data such as agent state, checkpoints, or reference information.
	- data -- Stores extra messages, structured, non-conversational data such as agent state, checkpoints, or reference information.


		Memory processing modes (controlled by the `infer` parameter):

		- Fact memory -- A processed representation of the message. The large language model (LLM) associated with the memory container extracts and stores key factual information or knowledge from the original text.

	- Fact memory -- A processed representation of the message. The large language model (LLM) associated with the memory container extracts and stores key factual information or knowledge from the original text.
	- infer=true -- Use large language model (LLM) to extract key information or knowledge from the messages.


		- Fact memory -- A processed representation of the message. The large language model (LLM) associated with the memory container extracts and stores key factual information or knowledge from the original text.

		- Raw message memory -- The unprocessed message content.


	- infer=false -- Only store raw messages and data in working memory.

	`messages` \| List \| Conditional \| A list of messages for conversation memory. Each message requires `content` and may include a `role` (commonly, `user` or `assistant`) when `infer` is set to `true`. Required for `memory_type` of `conversation`.
	`messages` \| List \| Conditional \| A list of messages for conversational payload. Each message requires `content` and may include a `role` (commonly, `user` or `assistant`) when `infer` is set to `true`. Required for `payload_type` of `conversational`.

	`structured_data` \| Object \| Conditional \| Structured data content for data memory. Required for `memory_type` of `data`.
	`structured_data` \| Map<String, Object> \| Conditional \| Structured data content for data memory. Required for `memory_type` of `data`.

	`memory_type` \| String \| Required \| The type of memory: `conversation` or `data`.
	`payload_type` \| String \| Required \| The type of payload: `conversational` or `data`.

	`namespace` \| Object \| Optional \| Namespace context for organizing memories (e.g., `user_id`, `session_id`, `agent_id`).
	`namespace` \| List<String> \| Optional \| Namespace context for organizing memories (e.g., `user_id`, `session_id`, `agent_id`).

	`metadata` \| Object \| Optional \| Additional metadata for the memory (e.g., `status`, `branch`, custom fields).
	`metadata` \| Map<String, String> \| Optional \| Additional metadata for the memory (e.g., `status`, `branch`, custom fields).

	`infer` \| Boolean \| Optional \| Controls whether the LLM infers context from messages. Default is `true` for conversation memory, `false` for data memory. When `true`, the LLM extracts factual information from the original text and stores it as the memory. When `false`, the memory contains the unprocessed message and you must explicitly specify the `role` in each message.
	`infer` \| Boolean \| Optional \| Controls whether use LLM to extract key information from messages. Default is `false`. When `true`, the LLM extracts key information from the original text and stores it as the memory.


		The request fields vary depending on the memory type being updated:

		### For session memory


		Use this API to create a memory container to hold agentic memories. The container can have two model types associated with it:

		- A text embedding model for vectorizing the message so it can be searched. Use a text embedding model for dense vector embeddings or a sparse encoding model for sparse vector formats. If no embedding model is specified, messages are stored but cannot be used for vector-based searches.


		### LLM

		Currently, agentic memory supports only the Anthropic Claude model for LLM capabilities. Starting with OpenSearch 3.3, this feature will be generally available and will include support for additional LLM providers, such as OpenAI.

Add comprehensive agentic memory API documentation for OpenSearch 3.3 #11169

Are you sure you want to change the base?

Add comprehensive agentic memory API documentation for OpenSearch 3.3 #11169

Uh oh!

Conversation

dhrubo-os commented Oct 1, 2025 • edited by kolchfa-aws Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Description

Issues Resolved

Version

Frontend features

Checklist

Uh oh!

github-actions bot commented Oct 1, 2025

Uh oh!

kolchfa-aws commented Oct 2, 2025

Uh oh!

kolchfa-aws commented Oct 9, 2025

Uh oh!

dhrubo-os commented Oct 10, 2025

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

ylwu-amzn Oct 10, 2025 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

ylwu-amzn Oct 10, 2025 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

ylwu-amzn Oct 10, 2025 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

dhrubo-os commented Oct 1, 2025 •

edited by kolchfa-aws

Loading

ylwu-amzn Oct 10, 2025 •

edited

Loading

ylwu-amzn Oct 10, 2025 •

edited

Loading

ylwu-amzn Oct 10, 2025 •

edited

Loading

ylwu-amzn Oct 10, 2025 •

edited

Loading

ylwu-amzn Oct 10, 2025 •

edited

Loading