SkillSpector · workspace-hetu

Dynamic __import__() can load arbitrary modules at runtime, bypassing static analysis and potentially importing malicious code.

scripts/deep_reasoning/calibrator.py

compile() creates code objects from strings. When combined with exec()/eval(), it enables obfuscated code execution.

scripts/deep_reasoning/embed_cal.py

subprocess module calls execute external commands. Without careful input validation, this enables command injection.

ziwei_engine.py

Skill attempts to fill the context window with filler content, displacing legitimate instructions and safety constraints. This can degrade agent performance or bypass safety boundaries.

daliuren_calc.py

Skill attempts to fill the context window with filler content, displacing legitimate instructions and safety constraints. This can degrade agent performance or bypass safety boundaries.

scripts/backtest_engine.py

Skill attempts to fill the context window with filler content, displacing legitimate instructions and safety constraints. This can degrade agent performance or bypass safety boundaries.

scripts/deep_reasoning/blind_school.py

Skill attempts to fill the context window with filler content, displacing legitimate instructions and safety constraints. This can degrade agent performance or bypass safety boundaries.

scripts/deep_reasoning/numerology_interpreter.py

Skill attempts to fill the context window with filler content, displacing legitimate instructions and safety constraints. This can degrade agent performance or bypass safety boundaries.

scripts/deep_reasoning/numerology_interpreter_v3bak.py

Skill attempts to fill the context window with filler content, displacing legitimate instructions and safety constraints. This can degrade agent performance or bypass safety boundaries.

scripts/deep_reasoning/triple_cross.py

Skill attempts to fill the context window with filler content, displacing legitimate instructions and safety constraints. This can degrade agent performance or bypass safety boundaries.

scripts/deep_reasoning/triple_cross.py

Skill attempts to fill the context window with filler content, displacing legitimate instructions and safety constraints. This can degrade agent performance or bypass safety boundaries.

scripts/deep_reasoning/triple_cross.py

Skill attempts to fill the context window with filler content, displacing legitimate instructions and safety constraints. This can degrade agent performance or bypass safety boundaries.

scripts/deep_reasoning/triple_cross.py

Skill attempts to fill the context window with filler content, displacing legitimate instructions and safety constraints. This can degrade agent performance or bypass safety boundaries.

scripts/deep_reasoning/triple_cross_v2.py

Skill attempts to fill the context window with filler content, displacing legitimate instructions and safety constraints. This can degrade agent performance or bypass safety boundaries.

scripts/deep_reasoning/triple_cross_v2.py

Model output is used without validation or sanitization. Unvalidated output injected into downstream contexts (SQL, shell, HTML) enables injection attacks and arbitrary code execution.

ziwei_engine.py

Skill modifies its own code, configuration, or behavior at runtime. Self-modification enables an agent to escalate privileges, disable safety constraints, or install persistent backdoors.

TOOLS.md

Dependencies lack version pinning, allowing potential malicious package updates. Consider pinning versions.

package.json