claude-cli-agent
Warn
Audited by Socket on Apr 3, 2026
1 alert found:
SecuritySecuritySKILL.md
MEDIUMSecurityMEDIUM
SKILL.md
SUSPICIOUS. The skill's main purpose aligns with using Claude CLI for delegated analysis, and the external service appears to be Anthropic's official tooling, not a credential-harvesting proxy. But the documentation is internally inconsistent about dependencies, omits the essential `claude` dependency, and explicitly advises `--dangerously-skip-permissions`, which expands data exposure risk during autonomous analysis of local content.
Confidence: 91%Severity: 71%
Audit Metadata