warm-integrations
Adapters that put WARM behind the memory, vector-store and retriever interfaces of Python agent and RAG frameworks, over the warm-server HTTP service.
Install
Framework dependency trees conflict: install exactly one extra per environment.
pip install "warm-integrations[langchain]" # or another extra from the table
warm-server --input-dim 384 # the vector length your embeddings produce
Quickstart
from langchain_core.documents import Document
from writerslogic_warm_integrations.core import HashingEmbedder
from writerslogic_warm_integrations.langchain import EmbedderAsLangChain, WarmVectorStore
store = WarmVectorStore(EmbedderAsLangChain(HashingEmbedder(384)), url="http://127.0.0.1:8080", api_key="dev", tenant_id="acme", collection="docs")
store.add_documents([Document(page_content="Rust has ownership and borrowing.", metadata={"topic": "code"})], ids=["rust"])
print(store.similarity_search("borrowing", k=4, filter={"topic": "code"}))
Run warm-server --no-auth --input-dim 384 (or set WARM_API_KEY=dev and pass api_key="dev") first. HashingEmbedder is lexical and offline; pass a real embedding model for semantic search. One runnable offline example per adapter is in examples/.
Options
Every adapter accepts the same WARM settings as keyword arguments or config=WarmConfig(...):
| Name | Type | Default | Meaning |
|---|---|---|---|
url, api_key, timeout, transport |
str, str, float, httpx transport | local server | Connection to warm-server |
tenant_id |
str | default tenant | Separate store per tenant |
collection |
str | none | Namespace inside a tenant |
embedder / the framework's embeddings |
Embedder |
HashingEmbedder(input_dim or 384) |
Encoder; text_queries=True lets a model-enabled server embed text |
max_fetch |
int | 100 | Result ceiling; match the server's --max-top-k |
overfetch |
int | 5 | Accepted for compatibility |
Adapters
| Extra | Host (tested version) | Page |
|---|---|---|
langchain |
langchain-core 1.6.12 | LangChain |
langgraph |
langgraph 1.2.16 | LangGraph |
llamaindex |
llama-index-core 0.14.25 | LlamaIndex |
haystack |
haystack-ai 3.4.0 | Haystack |
dspy |
dspy 3.4.1, 3.5.0 | DSPy |
openai-agents |
openai-agents 0.24.0 | OpenAI Agents |
pydantic-ai |
pydantic-ai-slim 2.58.0 | Pydantic AI |
smolagents |
smolagents 1.26.0 | smolagents |
litellm |
litellm 1.103.10 | LiteLLM |
crewai |
crewai 1.15.28 | CrewAI |
autogen |
autogen 0.7.5 | AutoGen |
ag2 |
ag2 1.1.2 | AutoGen and AG2 |
semantic-kernel |
semantic-kernel 1.45.0 | Semantic Kernel |
mem0 |
mem0ai 2.2.1 | mem0 |
letta |
letta-client 1.12.1 | Letta |
Compatibility
Python 3.11+ (tested 3.12), warm-sdk >= 0.1.3, warm-server of the same WARM release. Tests: WARM_SERVER_BIN=/path/to/warm-server ./run-matrix.sh [extra ...] (one isolated environment per extra).
Reference
Integrations overview, verification matrix and shared limits.
Metadata
Release files for warm-integrations 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 |
|---|---|---|---|---|
| warm_integrations-0.1.0.tar.gz | 82.7 kB | With provenance | Details for warm_integrations-0.1.0.tar.gz |
Built distribution (wheel)
| File | Interpreter | ABI | Platform | Actions Reset |
|---|---|---|---|---|
| warm_integrations-0.1.0-py3-none-any.whl | Python 3 | none | any | Details for warm_integrations-0.1.0-py3-none-any.whl |
Total release size: 196.6 kB
Release files / warm_integrations-0.1.0.tar.gz
| Download URL | warm_integrations-0.1.0.tar.gz |
|---|---|
| Size | 82.7 kB |
| Tags | Source |
|
SHA-256 checksum How to use checksums |
063c795eaefeedc5fbf287976af538d1937c260532a646471d02c999f79f8914
|
|
BLAKE2b-256 checksum How to use checksums |
73f0e20463cd346f2737014da981523e06fb87ba70941082a33c1bdefff191aa
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
Yes |
| Uploaded via |
twine/7.0.0 CPython/3.13.16
|
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.
PyPI Publish Attestation
PyPI verified that this artifact, at this checksum, originated from the publisher listed below.
Signed by GitHub Actions, verified by PyPI on Nov 19, 2026.
Transparency logRelease files / warm_integrations-0.1.0-py3-none-any.whl
| Download URL | warm_integrations-0.1.0-py3-none-any.whl |
|---|---|
| Size | 113.9 kB |
| Tags | Python 3 |
|
SHA-256 checksum How to use checksums |
6bdf49006ee359570ca593a46379c8eac36b1eec82a38508aba3d6ef7792ec4c
|
|
BLAKE2b-256 checksum How to use checksums |
077adaff4b799c336047d701c8659e2bcb88157577376952d5acb2d6f812030f
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
Yes |
| Uploaded via |
twine/7.0.0 CPython/3.13.16
|
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.
PyPI Publish Attestation
PyPI verified that this artifact, at this checksum, originated from the publisher listed below.
Signed by GitHub Actions, verified by PyPI on Nov 19, 2026.
Transparency logSecurity for warm-integrations 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.
Full provenance
Both files in this release have been uploaded with provenance metadata.
Attestation for warm_integrations-0.1.0.tar.gz
PyPI Publish Attestation
PyPI verified that this artifact, at this checksum, originated from the publisher listed below.
Signed by GitHub Actions, verified by PyPI on Nov 19, 2026.
Transparency logA 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.