agentic-workflow
Warn
Audited by Socket on Mar 6, 2026
1 alert found:
AnomalyAnomalySKILL.md
LOWAnomalyLOW
SKILL.md
The skill targets productivity and multi-agent orchestration, but it embeds risky patterns such as a curl|bash installer and unverified external tooling in its container/build context. This elevates supply-chain and data-exposure risk if adopted without safeguards. Recommended hardening includes replacing curl|sh with pinned installers or package managers with integrity checks, using secret vaults for API keys, enabling multi-stage builds, and auditing MCP tool access to minimize untrusted code execution. Overall, risk is moderate-to-high depending on usage.
Confidence: 68%Severity: 65%
Audit Metadata