coding-agent
Warn
Audited by Socket on Oct 8, 2026
1 alert found:
SecuritySecuritySKILL.md
MEDIUMSecurityMEDIUM
SKILL.md
SUSPICIOUS. Most workflows align with a coding-agent skill, and the named major tools are largely consistent with official distribution. However, the skill normalizes approval-bypass modes, enables autonomous GitHub posting/pushing, and includes a third-party Pi agent install that may receive prompts and API keys without strong provenance evidence. The footprint is broader and riskier than a simple background-runner guide.
Confidence: 88%Severity: 74%
Audit Metadata