ClawAudit verdict
model-distill-master
Reads files AND uses eval/Function
Model distillation workflow skill that downloads open-source models from ModelScope and uses pip-installed ML libraries; dynamic_eval and data_encoding are expected in ML training pipelines, no exfiltration present.
Automated static analysis — not a human review. ClawAudit flags capabilities, not confirmed intent, and can produce false positives. Disagree with this verdict? Use Dispute below.
What it does
These are capability combinations: each listed behavior occurs in the skill, but ClawAudit detects co-occurrence — it does not verify that one flows into another. Read the code to confirm a live chain.
Reads files AND uses eval/Function — may execute locally stored payloads
LLM05 · LLM06 · ASI05
Encodes data AND uses eval — the obfuscated-execution pattern (atob + eval; data-flow not verified)
LLM05 · ASI05 · ASI10
Permission integrity
file_read
package_install
Findings (5)
Uses eval() — can execute arbitrary code
SKILL.md · code · eval(
Pipe to python — executes piped content as Python code
.clawhub/publish-notes.md · prose · downgraded · | Python
apt-get install — installs system packages
scripts/download_model.sh · prose · downgraded · apt-get install
subprocess execution — runs system commands from Python
scripts/run_full_pipeline.py · prose · downgraded · subprocess.run(
Python shutil file operation — copies/moves/deletes files
scripts/package_deployment.py · prose · downgraded · shutil.copytree(
Why the tier is capped
Execution sink present in raw bytes (Hard Floor: class D). Final tier capped at Caution — cannot be lifted by any downgrade, example-payload opt-in, or allowlist.
Permissions & capabilities
No declared permissions — minimal attack surface.
package_installdynamic_evalfile_readdata_encoding Thanks — recorded.