crewai-memory-cosmosdb
An Azure Cosmos DB (NoSQL) StorageBackend for CrewAI's unified Memory — hierarchical scopes, categories, metadata filters, importance/recency, and native vector search via VectorDistance (diskANN index).
pip install crewai-memory-cosmosdb
from crewai import Crew
from crewai.memory import Memory
from crewai_memory_cosmosdb import CosmosDBMemoryBackend
backend = CosmosDBMemoryBackend(
endpoint="https://<acct>.documents.azure.com:443/", key="<key>",
database_name="crewai", container_name="memory", dimensions=3072, # match your embedder
)
memory = Memory(storage=backend)
crew = Crew(agents=[...], tasks=[...], memory=memory)
Or once for every Crew(memory=True): set_memory_storage_factory(lambda spec: CosmosDBMemoryBackend(...)).
How it works
- A container partitioned by
/scope, one document per record (id= record id), created with a vector embedding policy on/embedding(cosine,dimensions) and adiskANNvector index (vector_index_type="quantizedFlat"|"flat"to change). The policy is fixed at container creation — use a new container to changedimensions. searchis one SQL query:ORDER BY VectorDistance(c.embedding, @q)with the scope subtree inWHERE; the score is the cosine similarity.categories/metadata_filter/min_scoreare applied on the candidates.- Scope tree, category counts, listing,
delete(older_than=...),resetare cross-partition SQL queries. - The account needs Vector Search for NoSQL:
az cosmosdb update -n <acct> -g <rg> --capabilities EnableServerless EnableNoSQLVectorSearch(list all existing capabilities; propagation can take a few minutes).
Docs: https://skamalj.github.io/agentstate-reducer/ · part of crewai-memory
License
MIT
Metadata
Release files for crewai-memory-cosmosdb 0.1.0
For a detailed explanation of source distributions (sdists) and built distributions (wheels), please see the package formats documentation.
Source distribution (sdist)
| File | Size | Uploaded | Provenance | Actions |
|---|---|---|---|---|
| crewai_memory_cosmosdb-0.1.0.tar.gz | 4.3 kB | Without provenance | Details for crewai_memory_cosmosdb-0.1.0.tar.gz |
Built distribution (wheel)
| File | Interpreter | ABI | Platform | Actions Reset |
|---|---|---|---|---|
| crewai_memory_cosmosdb-0.1.0-py3-none-any.whl | Python 3 | none | any | Details for crewai_memory_cosmosdb-0.1.0-py3-none-any.whl |
Total release size: 8.6 kB
Release files / crewai_memory_cosmosdb-0.1.0.tar.gz
| Download URL | crewai_memory_cosmosdb-0.1.0.tar.gz |
|---|---|
| Size | 4.3 kB |
| Tags | Source |
|
SHA-256 checksum How to use checksums |
c12e9fbaf1cc4e08137aa7ffbf5a396283b6213568493c11bd6b83a5f1628eaa
|
|
BLAKE2b-256 checksum How to use checksums |
b14bf10f85032f00d8626f9b2a11e8dcdff80b41aa7b640009eba1a2f109ffb5
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
No |
| Uploaded via |
uv/0.11.23 {"installer":{"name":"uv","version":"0.11.23","subcommand":["publish"]},"python":null,"implementation":{"name":null,"version":null},"distro":{"name":"Ubuntu","version":"24.04","id":"noble","libc":null},"system":{"name":null,"release":null},"cpu":null,"openssl_version":null,"setuptools_version":null,"rustc_version":null,"ci":null}
|
Release files / crewai_memory_cosmosdb-0.1.0-py3-none-any.whl
| Download URL | crewai_memory_cosmosdb-0.1.0-py3-none-any.whl |
|---|---|
| Size | 4.3 kB |
| Tags | Python 3 |
|
SHA-256 checksum How to use checksums |
024cddce6df17208b7f604e6ac2e5e771125c895a6ca28af5618bb5b1d9618b0
|
|
BLAKE2b-256 checksum How to use checksums |
fff7381419fd3978d4ea0df1076fb59c26c12f32449106b46580080f56c8679d
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
No |
| Uploaded via |
uv/0.11.23 {"installer":{"name":"uv","version":"0.11.23","subcommand":["publish"]},"python":null,"implementation":{"name":null,"version":null},"distro":{"name":"Ubuntu","version":"24.04","id":"noble","libc":null},"system":{"name":null,"release":null},"cpu":null,"openssl_version":null,"setuptools_version":null,"rustc_version":null,"ci":null}
|
Security for crewai-memory-cosmosdb 0.1.0
Provenance
Provenance describes where a file came from. On PyPI, provenance is shared via attestations, which provide a verifiable record of the build or publishing details. View details, limitations and caveats.
No provenance found
None of the files in this release were uploaded with provenance metadata.
A note on package security
PyPI is a repository for distributing packages, not a service for auditing them. While third-party security researchers and our staff do their best to detect and remove malicious packages, PyPI cannot guarantee the quality, security or suitability of any given package.
To learn more about protecting yourself from malicious actors, visit our package security documentation.