← back to autoresearch-agent

Skill Scanner · autoresearch-agent

independent scanner by Cisco AI Defense · skill by DylanCkawalec · how it works ↗

WARNINGmax severity: HIGH

Shell command execution with shell=True enabled; CODE EXECUTION detected by YARA

scanned 2026-08-22

Findings (10)

HIGHcommand_injection

Shell command execution with shell=True enabled

evaluators\benchmark_size.py:26

HIGHcommand_injection

Shell command execution with shell=True enabled

evaluators\benchmark_size.py:34

MEDIUMcommand_injection

CODE EXECUTION detected by YARA

evaluators\benchmark_size.py:35

HIGHcommand_injection

Shell command execution with shell=True enabled

evaluators\benchmark_speed.py:20

HIGHcommand_injection

Shell command execution with shell=True enabled

evaluators\benchmark_speed.py:25

HIGHcommand_injection

Shell command execution with shell=True enabled

evaluators\build_speed.py:20

HIGHcommand_injection

Shell command execution with shell=True enabled

evaluators\build_speed.py:23

MEDIUMcommand_injection

CODE EXECUTION detected by YARA

evaluators\memory_usage.py:17

MEDIUMcommand_injection

CODE EXECUTION detected by YARA

evaluators\memory_usage.py:34

HIGHcommand_injection

Shell command execution with shell=True enabled

evaluators\test_pass_rate.py:13

What the verdicts mean

Skill Scanner reports on SkillMD's shared five-tier scale. See how Skill Scanner works ↗.

PASS

Reported as safe — no findings

CAUTION

Findings up to MEDIUM severity

WARNINGthis skill

Findings of HIGH severity

FAIL

Findings of CRITICAL severity

INCONCLUSIVE

Scan could not complete