atlassian-mcp

Warn

Audited by Socket on Apr 30, 2026

1 alert found:

Security
SecurityMEDIUM
SKILL.md

SUSPICIOUS. The skill’s stated purpose is coherent for Atlassian automation, but its concrete execution path is weaker than advertised: it runs an unpinned third-party MCP server via npx, the installer/example command does not cleanly match the upstream project’s documented path, and it forwards Jira/Confluence API tokens to that external code. No clear malicious exfiltration endpoint is present, so this is not confirmed malware, but the install-trust and credential-forwarding footprint is disproportionately risky for a documentation-style skill.

Confidence: 87%Severity: 83%
Audit Metadata
Analyzed At
Apr 30, 2026, 01:11 AM
Package URL
pkg:socket/skills-sh/jeffallan%2Fclaude-skills%2Fatlassian-mcp%2F@7dc373b304df7ad1382d7bb75e9cdbdf4cb6495d