atlassian-mcp
Warn
Audited by Snyk on Mar 7, 2026
Risk Level: MEDIUM
Full Analysis
MEDIUM W012: Unverifiable external dependency detected (runtime URL that controls agent).
- Potentially malicious external URL detected (high risk: 1.00). The skill's runtime configuration instructs launching an external MCP server (e.g., "npx @anthropic/mcp-atlassian" / "uvx mcp-atlassian" / "pip install mcp-atlassian"), which fetches and executes remote package code and is required at runtime to provide the agent's toolset (thereby directly controlling agent prompts and behavior).
Audit Metadata