ceos/unsloth
1
0
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity Actions

Studio: remove OpenEnv and other unused packages (#6585)

Browse source 
* Studio: drop OpenEnv and unused ExecuTorch/open_spiel install deps

* Studio: drop 8 more unused install deps from extras

* Studio: restore tomli<3.11 for kernels; tidy dep-cleanup comments and tests

* [pre-commit.ci] auto fixes from pre-commit.com hooks

for more information, see https://pre-commit.ci

* Studio: refresh scan-packages baseline for scipy _external + unsloth-zoo tests

scipy moved its vendored array_api_compat from scipy/_lib to
scipy/_external, so the four allowlisted array_api_compat __init__.py
entries stopped matching and resurfaced as unsuppressed CRITICAL
"Downloads and executes remote code" findings on all three pip
scan-packages shards (extras, hf-stack, studio). Add the _external
paths next to the existing _lib ones so both scipy layouts stay covered.

Allowlist two unsloth-zoo test-file false positives now present in the
hf-stack shard: tests/test_mlx_save_export_regressions.py (writes to
/tmp dropper) and tests/test_mlx_trainer_internals.py (obfuscation plus
exec/eval).

Drop nine stale entries for packages removed from the Studio
requirements and no longer in any shard closure (evaluate, pytest,
hypothesis, kgb, langid), confirmed absent via with-deps resolution of
all three shards.

---------

Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
Co-authored-by: Daniel Han <danielhanchen@gmail.com>
This commit is contained in:
oobabooga 2026-06-23 11:20:47 -03:00 • committed by GitHub
commit 1cc785e5a0
No known key found for this signature in database
GPG key ID: B5690EEEBB952194
8 changed files with 57 additions and 103 deletions
Download patch file Download diff file Expand all files Collapse all files

105
scripts/scan_packages_baseline.json
View file

@ -72,13 +72,6 @@
"severity": "CRITICAL",
"evidence": "Archive: L317: a['TarFileType'] = tarfile.open(fileobj=_fileW,mode='w')\nNetwork: L330: x['SocketType'] = _socket = socket.socket()"
},
{
"package": "evaluate",
"file": "evaluate/utils/file_utils.py",
"check": "C2 polling/beaconing loop detected",
"severity": "CRITICAL",
"evidence": "L261: while True:"
},
{
"package": "execnet",
"file": "execnet/gateway_base.py",
@ -401,27 +394,6 @@
"severity": "CRITICAL",
"evidence": "Base64: L488: decoded_bytes = base64.b64decode(base64_encoded)\nSubprocess: L80: return subprocess.call(['which', name], | L100: p = subprocess.Popen(['pbcopy', 'w'], | L105: p = subprocess.Popen(['pbpaste', 'r'],"
},
{
"package": "pytest",
"file": "_pytest/_py/path.py",
"check": "Downloads and executes remote code",
"severity": "CRITICAL",
"evidence": "L1153: exec(f.read(), mod.__dict__)"
},
{
"package": "pytest",
"file": "_pytest/capture.py",
"check": "Reverse shell / bind shell pattern",
"severity": "CRITICAL",
"evidence": "L483: os.dup2(self.targetfd_invalid, targetfd) | L522: os.dup2(self.tmpfile.fileno(), self.targetfd) | L532: os.dup2(self.targetfd_save, self.targetfd)"
},
{
"package": "pytest",
"file": "_pytest/config/__init__.py",
"check": "Reverse shell / bind shell pattern",
"severity": "CRITICAL",
"evidence": "L260: os.dup2(devnull, sys.stdout.fileno())"
},
{
"package": "python-dateutil",
"file": "dateutil/__init__.py",
@ -527,6 +499,34 @@
"severity": "CRITICAL",
"evidence": "L13: __import__(__package__ + '.linalg') | L14: __import__(__package__ + '.fft')"
},
{
"package": "scipy",
"file": "scipy/_external/array_api_compat/cupy/__init__.py",
"check": "Downloads and executes remote code",
"severity": "CRITICAL",
"evidence": "L12: __import__(__package__ + '.linalg') | L13: __import__(__package__ + '.fft')"
},
{
"package": "scipy",
"file": "scipy/_external/array_api_compat/dask/array/__init__.py",
"check": "Downloads and executes remote code",
"severity": "CRITICAL",
"evidence": "L16: __import__(__package__ + '.linalg') | L17: __import__(__package__ + '.fft')"
},
{
"package": "scipy",
"file": "scipy/_external/array_api_compat/numpy/__init__.py",
"check": "Downloads and executes remote code",
"severity": "CRITICAL",
"evidence": "L23: __import__(__package__ + \".linalg\") | L25: __import__(__package__ + \".fft\")"
},
{
"package": "scipy",
"file": "scipy/_external/array_api_compat/torch/__init__.py",
"check": "Downloads and executes remote code",
"severity": "CRITICAL",
"evidence": "L13: __import__(__package__ + '.linalg') | L14: __import__(__package__ + '.fft')"
},
{
"package": "sentencepiece",
"file": "sentencepiece/__init__.py",
@ -814,6 +814,13 @@
"severity": "CRITICAL",
"evidence": "L67: input_gguf=\"/tmp/in.gguf\","
},
{
"package": "unsloth-zoo",
"file": "tests/test_mlx_save_export_regressions.py",
"check": "Writes to /tmp and executes (staged dropper)",
"severity": "CRITICAL",
"evidence": "L164: temporary_location=\"/tmp/ignored\","
},
{
"package": "unsloth-zoo",
"file": "tests/test_upstream_pinned_symbols_transformers.py",
@ -933,13 +940,6 @@
"severity": "HIGH",
"evidence": "Key: L187: \"-----BEGIN PUBLIC KEY-----\", | L188: \"-----BEGIN RSA PUBLIC KEY-----\",\nNetwork: L225: http_client: httpx.AsyncClient | None = None, | L411: else httpx.AsyncClient(timeout=httpx.Timeout(10.0))"
},
{
"package": "hypothesis",
"file": "hypothesis/internal/scrutineer.py",
"check": "Anti-analysis/sandbox evasion + suspicious behavior",
"severity": "HIGH",
"evidence": "Anti: L76: return sys.gettrace() is None | L113: sys.settrace(self.trace) | L136: sys.settrace(None)"
},
{
"package": "ipython",
"file": "IPython/core/debugger.py",
@ -982,20 +982,6 @@
"severity": "HIGH",
"evidence": "Obfusc: L709: return compile(source, filename, \"exec\")\nExec: L1228: exec(code, namespace)"
},
{
"package": "kgb",
"file": "kgb/spies.py",
"check": "Advanced obfuscation (marshal/compile/zlib) + exec/eval",
"severity": "HIGH",
"evidence": "Obfusc: L934: eval(compile(func_code_str, '<string>', 'exec'),\nExec: L934: eval(compile(func_code_str, '<string>', 'exec'),"
},
{
"package": "langid",
"file": "langid/train/common.py",
"check": "Advanced obfuscation (marshal/compile/zlib) + exec/eval",
"severity": "HIGH",
"evidence": "Obfusc: L44: yield marshal.load(t)\nExec: L85: key = eval(row[0])"
},
{
"package": "matplotlib",
"file": "matplotlib/sphinxext/plot_directive.py",
@ -1108,20 +1094,6 @@
"severity": "HIGH",
"evidence": "Obfusc: L45: mod = __import__(module_name, None, None, ['__all__'])\nExec: L154: exec(f.read(), custom_namespace)"
},
{
"package": "pytest",
"file": "_pytest/_py/path.py",
"check": "Advanced obfuscation (marshal/compile/zlib) + exec/eval",
"severity": "HIGH",
"evidence": "Obfusc: L626: mod = __import__(hashtype) | L1118: __import__(modname)\nExec: L1153: exec(f.read(), mod.__dict__)"
},
{
"package": "pytest",
"file": "_pytest/assertion/rewrite.py",
"check": "Advanced obfuscation (marshal/compile/zlib) + exec/eval",
"severity": "HIGH",
"evidence": "Obfusc: L393: co = marshal.load(fp) | L395: trace(f\"_read_pyc({source}): marshal.load error {e}\")\nExec: L188: exec(co, module.__dict__)"
},
{
"package": "scikit-learn",
"file": "sklearn/externals/array_api_compat/torch/__init__.py",
@ -1318,6 +1290,13 @@
"severity": "HIGH",
"evidence": "Obfusc: L3078: module = __import__('transformers', fromlist=[model_class_name])\nExec: L2960: exec(f\"from transformers.modeling_utils import ({', '.join(functions)})\", locals(), globals()) | L3006: exec(save_pretrained, globals(), functions)"
},
{
"package": "unsloth-zoo",
"file": "tests/test_mlx_trainer_internals.py",
"check": "Advanced obfuscation (marshal/compile/zlib) + exec/eval",
"severity": "HIGH",
"evidence": "Obfusc: L430: assert ppl == pytest.approx(__import__(\"math\").exp(2.5))\nExec: L408: def eval(self):"
},
{
"package": "werkzeug",
"file": "werkzeug/routing/rules.py",