I keep getting this error no matter what I do: ImportError: cannot import name 'EnvVar' from 'crewai.tools'

[AImicrowave]

Hi,
I’m trying to install and run CrewAI Studio on my Mac, but I keep running into dependency issues with the crewai and crewai-tools packages. The versions specified in the README (crewai==0.22.6 and crewai-tools==0.0.51) are not available on PyPI, and the latest available versions are not compatible with each other or with CrewAI Studio’s codebase.
Specifically, I get this error:
ImportError: cannot import name 'EnvVar' from 'crewai.tools'
It seems that the latest crewai_tools package expects an EnvVar symbol in crewai.tools, but it does not exist in the latest crewai package.
Could you please provide:
A working requirements.txt or list of package versions that are known to work together?
Or, if possible, the old .whl files for crewai==0.22.6 and crewai-tools==0.0.51?
Or, ideally, an updated version of CrewAI Studio that works with the latest crewai and crewai-tools?
Thank you so much for your help and for making this project open source!

[strnad]

Hi, are you sure you are using this repository and not some fork of it? I didn't put any version specifications neighter to readme or requirements.txt. Are you using venv, conda or docker? What is your python version?

[AImicrowave]

Hi!
Thank you for your quick response.
I am using the official CrewAI Studio repository (not a fork).
I set up a virtual environment using venv.
My Python version is 3.11.
I installed the dependencies using pip install -r requirements.txt and then installed crewai and crewai-tools from PyPI (since there are no version specs in the repo).
When I run the app, I get this error:
ImportError: cannot import name 'EnvVar' from 'crewai.tools'
(/Users/"my name"/Downloads/CrewAI-Studio-main/venv/lib/python3.11/site-packages/crewai/tools/init.py)
It seems the latest crewai_tools package expects an EnvVar symbol in crewai.tools, but it doesn’t exist in the latest crewai package.
Do you have a list of package versions that are known to work, or any advice on which versions of crewai and crewai-tools I should use?
Thank you very much.

[AImicrowave]

I didn't try using conda yet so i'm going to try doing it with conda (I have an older mac so Docker doesn't work)

[strnad]

maybe try to install it using pip without cache (--no-cache), or it could be because of python version... I have these packages and it works without any problem:

PS C:\CrewAI-Studio> python --version
Python 3.12.10
PS C:\CrewAI-Studio> & C:\CrewAI-Studio\venv\Scripts\Activate.ps1
(venv) PS C:\CrewAI-Studio> .\venv\Scripts\activate
(venv) PS C:\CrewAI-Studio> pip freeze
aiohappyeyeballs==2.6.1
aiohttp==3.12.13
aiosignal==1.4.0
alembic==1.16.2
altair==5.5.0
annotated-types==0.7.0
anthropic==0.57.1
anyio==4.9.0
appdirs==1.4.4
asgiref==3.9.0
asn1crypto==1.5.1
asttokens==3.0.0
attrs==25.3.0
babel==2.17.0
backoff==2.2.1
backrefs==5.9
bcrypt==4.3.0
beautifulsoup4==4.13.4
blinker==1.9.0
boto3==1.39.3
botocore==1.39.3
build==1.2.2.post1
cachetools==5.5.2
certifi==2025.6.15
cffi==1.17.1
cfgv==3.4.0
charset-normalizer==3.4.2
chroma-hnswlib==0.7.6
chromadb==0.5.23
click==8.2.1
cohere==5.15.0
colorama==0.4.6
coloredlogs==15.0.1
crewai==0.140.0
crewai-tools==0.49.0
cryptography==45.0.5
dataclasses-json==0.6.7
decorator==5.2.1
deprecation==2.1.0
dill==0.4.0
distlib==0.3.9
distro==1.9.0
docker==7.1.0
docling==2.40.0
docling-core==2.40.0
docling-ibm-models==3.7.0
docling-parse==4.1.0
docstring_parser==0.16
duckduckgo_search==8.1.1
durationpy==0.10
easyocr==1.7.2
embedchain==0.1.128
et_xmlfile==2.0.0
executing==2.2.0
fastapi==0.115.14
fastavro==1.11.1
filelock==3.18.0
filetype==1.2.0
flatbuffers==25.2.10
frozenlist==1.7.0
fsspec==2025.5.1
ghp-import==2.1.0
gitdb==4.0.12
GitPython==3.1.44
google-auth==2.40.3
googleapis-common-protos==1.70.0
gptcache==0.1.44
greenlet==3.2.3
groq==0.29.0
grpcio==1.73.1
h11==0.16.0
h2==4.2.0
hpack==4.1.0
httpcore==1.0.9
httptools==0.6.4
httpx==0.28.1
httpx-sse==0.4.0
huggingface-hub==0.33.2
humanfriendly==10.0
hyperframe==6.1.0
identify==2.6.12
idna==3.10
imageio==2.37.0
importlib_metadata==8.7.0
importlib_resources==6.5.2
instructor==1.9.0
ipython==9.4.0
ipython_pygments_lexers==1.1.1
jedi==0.19.2
Jinja2==3.1.6
jiter==0.10.0
jmespath==1.0.1
json5==0.12.0
json_repair==0.25.2
jsonlines==3.1.0
jsonpatch==1.33
jsonpickle==4.1.1
jsonpointer==3.0.0
jsonref==1.1.0
jsonschema==4.24.0
jsonschema-specifications==2025.4.1
kubernetes==33.1.0
lancedb==0.24.0
langchain==0.3.26
langchain-anthropic==0.3.17
langchain-cohere==0.3.5
langchain-community==0.3.27
langchain-core==0.3.68
langchain-experimental==0.3.4
langchain-groq==0.3.5
langchain-ollama==0.3.3
langchain-openai==0.2.14
langchain-text-splitters==0.3.8
langsmith==0.3.45
latex2mathml==3.78.0
lazy_loader==0.4
litellm==1.72.6
lxml==5.4.0
Mako==1.3.10
Markdown==3.8.2
markdown-it-py==3.0.0
marko==2.1.4
MarkupSafe==3.0.2
marshmallow==3.26.1
matplotlib-inline==0.1.7
mdurl==0.1.2
mem0ai==0.1.114
mergedeep==1.3.4
mkdocs==1.6.1
mkdocs-get-deps==0.2.0
mkdocs-material==9.6.15
mkdocs-material-extensions==1.3.1
mmh3==5.1.0
monotonic==1.6
mpire==2.10.2
mpmath==1.3.0
multidict==6.6.3
multiprocess==0.70.18
mypy_extensions==1.1.0
narwhals==1.46.0
networkx==3.5
ninja==1.11.1.4
nodeenv==1.9.1
numpy==2.2.6
oauthlib==3.3.1
ollama==0.5.1
onnxruntime==1.22.0
openai==1.93.0
opencv-python-headless==4.12.0.88
openpyxl==3.1.5
opentelemetry-api==1.34.1
opentelemetry-exporter-otlp-proto-common==1.34.1
opentelemetry-exporter-otlp-proto-grpc==1.34.1
opentelemetry-exporter-otlp-proto-http==1.34.1
opentelemetry-instrumentation==0.55b1
opentelemetry-instrumentation-asgi==0.55b1
opentelemetry-instrumentation-fastapi==0.55b1
opentelemetry-proto==1.34.1
opentelemetry-sdk==1.34.1
opentelemetry-semantic-conventions==0.55b1
opentelemetry-util-http==0.55b1
orjson==3.10.18
overrides==7.7.0
packaging==24.2
paginate==0.5.7
pandas==2.3.0
parso==0.8.4
pathspec==0.12.1
pdfminer.six==20250506
pdfplumber==0.11.7
pillow==11.3.0
platformdirs==4.3.8
pluggy==1.6.0
portalocker==2.10.1
posthog==3.25.0
pre_commit==4.2.0
primp==0.15.0
prompt_toolkit==3.0.51
propcache==0.3.2
protobuf==5.29.5
psycopg2-binary==2.9.10
pure_eval==0.2.3
pyarrow==20.0.0
pyasn1==0.6.1
pyasn1_modules==0.4.2
pyclipper==1.3.0.post6
pycparser==2.22
pydantic==2.11.7
pydantic-settings==2.10.1
pydantic_core==2.33.2
pydeck==0.9.1
Pygments==2.19.2
PyJWT==2.10.1
pylatexenc==2.10
pymdown-extensions==10.16
pyOpenSSL==25.1.0
pypdf==5.7.0
pypdfium2==4.30.1
PyPika==0.48.9
pyproject_hooks==1.2.0
pyreadline3==3.5.4
pyright==1.1.402
pysbd==0.3.4
python-bidi==0.6.6
python-dateutil==2.9.0.post0
python-docx==1.2.0
python-dotenv==1.1.1
python-pptx==1.0.2
pytube==15.0.0
pytz==2025.2
pyvis==0.3.2
pywin32==310
PyYAML==6.0.2
pyyaml_env_tag==1.1
qdrant-client==1.14.3
referencing==0.36.2
regex==2024.11.6
requests==2.32.4
requests-oauthlib==2.0.0
requests-toolbelt==1.0.0
rich==13.9.4
rpds-py==0.26.0
rsa==4.9.1
rtree==1.4.0
s3transfer==0.13.0
safetensors==0.5.3
schema==0.7.7
scikit-image==0.25.2
scipy==1.16.0
semchunk==2.2.2
setuptools==80.9.0
shapely==2.1.1
shellingham==1.5.4
six==1.17.0
smmap==5.0.2
sniffio==1.3.1
snowflake-connector-python==3.16.0
sortedcontainers==2.4.0
soupsieve==2.7
SQLAlchemy==2.0.41
stack-data==0.6.3
starlette==0.46.2
streamlit==1.46.1
sympy==1.14.0
tabulate==0.9.0
tenacity==9.1.2
tifffile==2025.6.11
tiktoken==0.9.0
tokenizers==0.20.3
toml==0.10.2
tomli==2.2.1
tomli_w==1.2.0
tomlkit==0.13.3
torch==2.7.1
torchvision==0.22.1
tornado==6.5.1
tqdm==4.67.1
traitlets==5.14.3
transformers==4.46.3
typer==0.16.0
types-requests==2.32.4.20250611
typing-inspect==0.9.0
typing-inspection==0.4.1
typing_extensions==4.14.1
tzdata==2025.2
urllib3==2.5.0
uv==0.7.19
uvicorn==0.35.0
virtualenv==20.31.2
watchdog==6.0.0
watchfiles==1.1.0
wcwidth==0.2.13
websocket-client==1.8.0
websockets==15.0.1
wrapt==1.17.2
xlsxwriter==3.2.5
yarl==1.20.1
zipp==3.23.0
zstandard==0.23.0

[AImicrowave]

Thank you for helping me, I will try using these versions and I will text here if it was succesful

[AImicrowave]

Hi again,
Thank you for your previous help and for sharing your working pip freeze output. I’ve tried to match your environment as closely as possible (using Python 3.12.1, crewai==0.140.0, crewai-tools==0.49.0, and the rest of your package versions).
However, I keep running into a dependency conflict that pip cannot resolve:
crewai==0.140.0 requires onnxruntime==1.22.0, but that version is not available for my platform (macOS, Python 3.12.1).
chromadb==0.5.23 requires onnxruntime>=1.14.1, which would work with the available versions, but not with the strict pin to 1.22.0 from crewai. Pip gives this error: ERROR: Cannot install -r requirements.txt (line 29) and -r requirements.txt (line 34) because these package versions have conflicting dependencies.
The conflict is caused by:
chromadb 0.5.23 depends on onnxruntime>=1.14.1
crewai 0.140.0 depends on onnxruntime==1.22.0
ERROR: ResolutionImpossible
I’ve tried removing the version pin for onnxruntime, but then crewai still requires 1.22.0 specifically, which isn’t available for my system.
Is there a workaround for this, or could you suggest a set of versions (or a requirements file) that would work on macOS? Or is there a way to relax the onnxruntime version requirement in crewai?
Thank you so much for your help and for maintaining this project!

[onestardao]

The problem in issue #98 is a version compatibility and missing symbol error.
The latest crewAI_tools package expects an EnvVar symbol in crewai.tools, but that symbol no longer exists in the current package version.

This means:

• The versions listed in the README (crewAI = 0.22.6 and crewAI_tools = 0.5.1) are either not available or incompatible.
• Installing crewAI_tools from PyPI without pinning the version results in ImportError: cannot import name 'EnvVar' from 'crewAI.tools'.
• The error will persist unless a matching pair of crewAI and crewAI_tools versions is used, or the code is updated to remove the dependency on EnvVar.

[RHuebner1972]

Using Windows 11. CrewAI Studio installed in d:\code\CrewAI-Studio.
System python: 3.12. (in C:\Python312

Using pip / venv type of environment (as opposed to conda or docker).
Ran install_venv.bat. No issues that I can see.

Ran through install, then run the run_venv.bat file. I am getting this error:

D:\Code\CrewAI-Studio\venv\Lib\site-packages\litellm\litellm_core_utils\core_helpers
.py:8 in

  5 import httpx
  6
  7 from litellm._logging import verbose_logger

❱ 8 from litellm.types.llms.openai import AllMessageValues
9
10 if TYPE_CHECKING:
11 │ from opentelemetry.trace import Span as _Span

D:\Code\CrewAI-Studio\venv\Lib\site-packages\litellm\types\llms\openai.py:36 in

  33 )
  34 from openai.types.embedding import Embedding as OpenAIEmbedding
  35 from openai.types.fine_tuning.fine_tuning_job import FineTuningJob

❱ 36 from openai.types.responses.response import (
37 │ IncompleteDetails,
38 │ Response,
39 │ ResponseOutputItem,
────────────────────────────────────────────────────────────────────────────────────────
ImportError: cannot import name 'ResponseTextConfig' from
'openai.types.responses.response'
(D:\Code\CrewAI-Studio\venv\Lib\site-packages\openai\types\responses\response.py)
Stopping...

[joaquinariasco-lab]

Hi,

Reading through this thread, this is a very real example of the kind of ecosystem friction a lot of agent developers are hitting right now, Python versions, pinned deps, broken symbols, platform-specific issues, etc.

I’m working on a small open agent interoperability layer that intentionally avoids heavy framework dependencies. The idea is not to replace CrewAI or AutoGen, but to let agents connect and exchange tasks/messages via a very thin interface, even if they come from different setups.

If you’re interested in experimenting (no pressure, it’s early-stage), here’s the repo:
https://github.com/joaquinariasco-lab/flowing.git

You can run two agents locally in minutes (Flask + simple REST), no complex dependency graph, and see agents communicate directly. I’d genuinely value feedback from someone who’s dealing with these real-world integration issues.

Totally optional, just sharing in case it’s useful.

Best,
Joaquin