build-iterated-agentic-loop
Warn
Audited by Socket on Jun 23, 2026
2 alerts found:
AnomalySecurityAnomalySKILL.md
LOWAnomalyLOW
SKILL.md
SUSPICIOUS: the skill's purpose is coherent, but it enables autonomous CI agents to make code changes and open PRs on a schedule, and the crucial install/auth behavior is deferred to missing reference files. The main concern is autonomy plus hidden downstream agent/runtime details, not confirmed malware or clear credential theft.
Confidence: 84%Severity: 66%
Securityreferences/agent-runner-templates.md
MEDIUMSecurityMEDIUM
references/agent-runner-templates.md
Audit Metadata