autoresearch
Warn
Audited by Socket on Sep 9, 2026
1 alert found:
AnomalyAnomalySKILL.md
LOWAnomalyLOW
SKILL.md
SUSPICIOUS: the stated purpose matches a research-orchestration skill, and the explicit package installs are normal PyPI usage, so there is no strong malware signal. However, the skill materially increases risk by mandating autonomous recurring execution, allowing ongoing actions without confirmation, encouraging outbound messaging/report delivery, and delegating into additional skills whose trust boundary is outside this skill. The footprint is coherent with autonomous research management but broader and riskier than a typical guide skill.
Confidence: 89%Severity: 58%
Audit Metadata