atlassian-mcp

MEDIUM W011: Third-party content exposure detected (indirect prompt injection risk).

• Third-party content exposure detected (high risk: 0.75). The required workflow includes querying Jira/Confluence and then parsing returned issue/page text (e.g., jira_get_issue, confluence_get_page, jira_search, confluence_search) which is outsider-authored content from other users, and that text can be fed into the agent’s LLM context via the tool results.

MEDIUM W012: Unverifiable external dependency detected (runtime URL that controls agent).

• Potentially malicious external URL detected (high risk: 0.90). The skill configuration launches external packages with npx (e.g., "@sooperset/mcp-atlassian" / "npx @anthropic/mcp-atlassian"), which will fetch and execute remote code at runtime and is therefore a high-confidence runtime-executed external dependency.