AzureCosmosDB · aayush3011 · Jul 11, 2026 · Jul 1, 2026 · Jul 1, 2026 · Jul 1, 2026
diff --git a/.env.template b/.env.template
@@ -18,15 +18,6 @@ COSMOS_DB_SUMMARIES_CONTAINER="memories_summaries"
 COSMOS_DB_TURNS_CONTAINER="memories_turns"
 COSMOS_DB_COUNTERS_CONTAINER=counter
 COSMOS_DB_LEASE_CONTAINER=leases
-# Throughput mode for all required Cosmos DB containers created by the toolkit
-# (memories, counter, and lease).
-# - serverless: default. The toolkit does not send container RU/s settings.
-#   Use this only with a Cosmos DB account configured for serverless.
-# - autoscale: the toolkit provisions all required containers with autoscale
-#   throughput using COSMOS_DB_AUTOSCALE_MAX_RU as the max RU/s cap.
-#   Default max RU/s is 1000.
-COSMOS_DB_THROUGHPUT_MODE=serverless
-COSMOS_DB_AUTOSCALE_MAX_RU=1000
 
 # ---- Processing thresholds (set to 0 to disable) ----
 THREAD_SUMMARY_EVERY_N=10
@@ -58,14 +49,6 @@ AI_FOUNDRY_ENDPOINT=https://<your-account>.openai.azure.com/
 AI_FOUNDRY_API_KEY=
 AI_FOUNDRY_EMBEDDING_DEPLOYMENT_NAME=text-embedding-3-large
 AI_FOUNDRY_EMBEDDING_DIMENSIONS=1536
-AI_FOUNDRY_EMBEDDING_DATA_TYPE=float32
-AI_FOUNDRY_EMBEDDING_DISTANCE_FUNCTION=cosine
-# Optional. Vector index type for the memories container: quantizedFlat
-# (default), diskANN, or flat. quantizedFlat works on any Cosmos DB account
-# (including the classic emulator); diskANN requires the DiskANN capability on
-# the Cosmos DB account, so opt into it explicitly when available.
-AI_FOUNDRY_EMBEDDING_VECTOR_INDEX_TYPE=quantizedFlat
-COSMOS_DB_FULL_TEXT_LANGUAGE=en-US
 
 # Embed raw conversation turns on write so they can be vector-searched via
 # search_turns(). The turns container is always provisioned with a

diff --git a/CHANGELOG.md b/CHANGELOG.md
@@ -1,6 +1,17 @@
 ## Release History
 
-## [Unreleased]
+## [0.3.0b1] (Unreleased)
+
+#### Features Added
+* Write-time in-place deduplication: near-duplicate memories fold into the existing record (same id, newer content) instead of creating a new doc. See [PR:#26](https://github.com/AzureCosmosDB/AgentMemoryToolkit/pull/26)
+* Reconciliation now resolves contradictions only, soft-deleting the loser with `superseded_by`; `get_memory_history()` walks that chain. See [PR:#26](https://github.com/AzureCosmosDB/AgentMemoryToolkit/pull/26)
+
+#### Bugs Fixed
+* Fixed a re-extraction loop that re-extracted the whole conversation every cycle (turns were never stamped `extracted_at` when vector dedup was on). See [PR:#26](https://github.com/AzureCosmosDB/AgentMemoryToolkit/pull/26)
+
+#### Other Changes
+* Reworked the extraction prompt (anti-inference, preserve specifics, topic-grouped memories) and simplified the schema to `fact`/`episodic` with fixed fact categories. See [PR:#26](https://github.com/AzureCosmosDB/AgentMemoryToolkit/pull/26)
+* Token-bounded extraction batches with per-batch failure isolation; embedding inputs truncated to the model token budget. See [PR:#26](https://github.com/AzureCosmosDB/AgentMemoryToolkit/pull/26)
 
 ## [0.2.0b3] (2026-07-08)
 
@@ -17,7 +28,6 @@
   `DEDUP_EVERY_N`, `THREAD_SUMMARY_EVERY_N`, `USER_SUMMARY_EVERY_N`); any key not
   present falls back to the environment/defaults, and `None` preserves today's
   env-only behavior. See [PR:#29](https://github.com/AzureCosmosDB/AgentMemoryToolkit/pull/29)
-
 ## [0.2.0b2] (2026-07-01)
 
 #### Features Added

diff --git a/Docs/azure_testing.md b/Docs/azure_testing.md
@@ -6,13 +6,13 @@ This guide covers the minimum Azure resources, deployment steps, throughput sett
 
 ## Required Azure Services
 
-| Service | Purpose |
-|---------|---------|
-| **Azure Cosmos DB for NoSQL** | Persistent memory store with vector and full-text indexes |
-| **Azure OpenAI / AI Services** | Embeddings and chat generation |
-| **Azure Functions** | Durable Functions orchestrator and activities |
-| **Azure Storage Account** | Required by Azure Functions |
-| **Application Insights** | Recommended for monitoring |
+| Service                        | Purpose                                                   |
+|--------------------------------|-----------------------------------------------------------|
+| **Azure Cosmos DB for NoSQL**  | Persistent memory store with vector and full-text indexes |
+| **Azure OpenAI / AI Services** | Embeddings and chat generation                            |
+| **Azure Functions**            | Durable Functions orchestrator and activities             |
+| **Azure Storage Account**      | Required by Azure Functions                               |
+| **Application Insights**       | Recommended for monitoring                                |
 
 ---
 
@@ -277,7 +277,7 @@ Bring the environment up in this order:
 9. write a few turns and verify a thread `summary` memory appears
 10. write more turns and verify `fact`, `procedural`, and `episodic` memories appear
 11. verify a per-user `user_summary` memory appears once `USER_SUMMARY_EVERY_N` turns have accumulated for that user
-12. test deduplication by writing two near-duplicate facts and confirming the dedup orchestrator merges them
+12. test deduplication by writing two near-duplicate facts and confirming the second folds into the first in place (one active record remains, its `updated_at` bumped) rather than creating a duplicate
 
 This keeps failures isolated and easier to diagnose.
 
@@ -367,15 +367,15 @@ az functionapp log tail \
 
 Common issues:
 
-| Symptom | Likely Cause |
-|---------|--------------|
-| 401 / 403 from Cosmos DB | Missing Cosmos DB RBAC |
-| 401 / 403 from Azure OpenAI | Missing OpenAI RBAC |
-| Durable Function starts but fails | Missing app settings or downstream RBAC |
-| `No memories found` | No turn memories exist, or all candidate turns predate the existing summary |
-| Search is slow | Embedding latency, index choice, or region mismatch |
-| Change feed trigger not firing | Verify `COSMOS_DB__accountEndpoint` is set and the function can write to the configured `COSMOS_DB_COUNTERS_CONTAINER` container |
-| Auto-processing not starting | Check threshold settings are > 0 in Function App configuration |
+| Symptom                           | Likely Cause                                                                                                                     |
+|-----------------------------------|----------------------------------------------------------------------------------------------------------------------------------|
+| 401 / 403 from Cosmos DB          | Missing Cosmos DB RBAC                                                                                                           |
+| 401 / 403 from Azure OpenAI       | Missing OpenAI RBAC                                                                                                              |
+| Durable Function starts but fails | Missing app settings or downstream RBAC                                                                                          |
+| `No memories found`               | No turn memories exist, or all candidate turns predate the existing summary                                                      |
+| Search is slow                    | Embedding latency, index choice, or region mismatch                                                                              |
+| Change feed trigger not firing    | Verify `COSMOS_DB__accountEndpoint` is set and the function can write to the configured `COSMOS_DB_COUNTERS_CONTAINER` container |
+| Auto-processing not starting      | Check threshold settings are > 0 in Function App configuration                                                                   |
 
 Recommended checks: