Skip to content

Security: deepwired/mcp-jira

Security

SECURITY.md

Security Policy

Token Safety

This MCP server handles Atlassian API tokens. The following measures are in place:

  • Tokens are never logged — the sanitizeError function redacts tokens from all error messages
  • Tokens are never written to files — configuration is read from environment variables only
  • Scope enforcement — the server blocks API calls outside the configured scopes before they reach Atlassian
  • Delete safety guardjira_delete_issue requires an explicit confirm: true parameter
  • Read-only by default — if JIRA_SCOPES is not set, only read tools are available
  • Pre-commit hook — the repo includes a git hook (.git/hooks/pre-commit) that blocks commits containing token patterns (ATATT tokens, Bearer values, hardcoded JIRA_API_TOKEN assignments). The only allowlisted dummy values are the exact strings 'token' and 'secret', used in test fixtures. Any other value will block the commit.

Reporting a Vulnerability

If you discover a security vulnerability in this project, please do not open a public issue.

Instead, please report it privately:

  1. Email the maintainer directly, or
  2. Use GitHub Security Advisories to report privately

We will respond within 48 hours and work with you to understand and address the issue.

Supported Versions

Version Supported
1.x Yes

Scope

This policy covers the mcp-jira-scoped npm package and the source code in this repository. It does not cover Atlassian's APIs or infrastructure.

There aren’t any published security advisories