autoresearch

Warn

Audited by Socket on Mar 16, 2026

1 alert found:

Security
SecurityMEDIUM
SKILL.md

SUSPICIOUS: the skill's purpose matches local autonomous optimization, but its footprint is high-risk because it grants persistent, unsupervised shell execution and repo mutation with explicit 'never stop' behavior. No clear credential theft, exfiltration, or external supply-chain abuse is present, so this is not confirmed malware, but it is a dangerous autonomous coding skill.

Confidence: 88%Severity: 74%
Audit Metadata
Analyzed At
Mar 16, 2026, 03:41 AM
Package URL
pkg:socket/skills-sh/dabiggm0e%2Fautoresearch-opencode%2Fautoresearch%2F@2cc9d87c25f86a3abf87cd304d2c1f4af65abf8e