Security and incident response
Report a vulnerability or incident: email security@dceniac.com with the subject "Security report".
We acknowledge security reports within 2 business days, and sooner for anything that looks critical.
What is in scope
Dceniac Markdown Writer for Jira and Spreadsheet & Table Field for Jira are Atlassian Forge apps ("Runs on Atlassian"). They execute on Atlassian's managed infrastructure, enforce least-privilege Jira scopes, and do not store or transmit customer content anywhere outside Atlassian. Specifically:
- Dceniac Markdown Writer for Jira calls Jira APIs exclusively as the signed-in user (
read:jira-work,write:jira-work,read:jira-user) and stores descriptions, comments, and rendered diagrams natively on the Jira issue. Unsaved drafts stay only in the user's own browser storage, scoped by site, account and issue, and expire after 14 days. - Spreadsheet & Table Field for Jira uses a single read-only scope (
read:custom-field-contextual-configuration:jira) to load custom field schemas owned by the app, evaluates formulas with a deterministic zero-evalparser, neutralizes spreadsheet formula injection on CSV/TSV exports, and stores all table configurations and issue values natively inside Jira.
The only infrastructure we operate ourselves is this website, a static site on Cloudflare that hosts documentation and legal policies and holds no customer data.
Because of that, the incidents this plan is written for are mainly: a vulnerability in an app's code or its dependencies, a compromised developer account (Atlassian, GitHub, Google or Cloudflare), and a malicious or broken app release.
Severity and fix times
We rate vulnerabilities with CVSS and fix them within the timeframes of Atlassian's Marketplace security bug fix policy:
| Severity | CVSS | Fixed within |
|---|---|---|
| Critical | 9.0 and above | 10 days |
| High | 7.0 to 8.9 | 4 weeks |
| Medium | 4.0 to 6.9 | 12 weeks |
| Low | below 4.0 | 25 weeks |
Incident response steps
This follows Atlassian's app security incident management guidelines.
- Triage. Confirm the report, rate its severity, and work out what is affected: which app and versions, since when, which customers, and whether any customer data was exposed.
- Notify Atlassian within 24 hours of confirming a security incident, with a P1 ticket in the Atlassian Ecosystem service desk. The ticket gives the scope, affected data and customers, containment steps, expected resolution time and a contact, and we add what we learn as we learn it.
- Contain. Depending on the cause: ship a fixed version, roll back to a known good version, revoke and rotate credentials and sessions on a compromised account, or ask Atlassian to restrict or delist the affected app until it is safe. While the incident is open we update Atlassian at least every 6 hours.
- Notify customers within 72 hours of identifying an incident that affects them, using Atlassian's incident communication template, through the Marketplace contacts of affected sites and our support email.
- Fix the cause. Release the permanent fix and check that it closes the issue.
- Review. Confirm nothing compromised remains, write down the root cause and what we changed to prevent a repeat, and share the review on the Atlassian ticket.
Day-to-day practices
- Multi-factor authentication on the accounts used to build and publish our apps.
- Unique, generated passwords kept in a password manager.
- Every change is reviewed and tested by the developer before it is deployed to production.
- Dependencies are checked with
npm auditbefore each release, and known issues are fixed or tracked. - Our apps log no customer content, credentials or tokens.