Coordinated vulnerability disclosure policy
The CVD policy of Arkitekt.
1. Scope
This policy covers vulnerabilities in:
- products: software that Arkitekt develops and distributes, and
- infrastructure: services and systems that Arkitekt operates, such as the hosts that publish this page.
We consider a report a valid vulnerability if it
- affects one of our products or our infrastructure, and
- preferably concerns information that is not yet public.
Output of automated tools or scanners without supporting explanation may not be treated as a valid report. We still read reports about issues that are already fixed, even when they don't start a new CVD process.
2. How to report
Use one of these contact options. Email addresses and telephone numbers are both accepted as ways for us to reach you; leave one with your report.
Product security (PSIRT)
For vulnerabilities in software Arkitekt ships.
Email: psirt@arkitekt.live
OpenPGP key: https://go.arkitekt.live/.well-known/pgp/psirt.asc
Fingerprint: D7B9 A162 DA4B C871 1C3C 9D46 EAF6 8605 EF40 FEA6
Infrastructure security (CSIRT)
For vulnerabilities in services and systems Arkitekt operates.
Email: csirt@arkitekt.live
OpenPGP key: https://go.arkitekt.live/.well-known/pgp/csirt.asc
Fingerprint: E904 021F 2EF9 5EC7 1A32 9A18 91EB 2A60 9F05 5969
You can also use the web form, with or without contact details. For confidential information we recommend encrypted and digitally signed email (OpenPGP).
A good report names the affected product or service and its version, describes the vulnerability and its impact, and includes steps to reproduce it. Complex issues may need further explanation or documentation from you. Good communication matters: if you don't answer technical or content questions, we may only be able to process your report in part, or not at all. Questions from you about the status of your report are always welcome.
We accept reports in: en, de.
Anonymous reports
You can report without giving any contact details through the web form. The form stores only what you type in, not your IP address. Because we can't ask you anything, an anonymous report can only be processed to a limited extent, or possibly not at all.
3. What we promise you
- Confidentiality. We treat every report confidentially, as far as the law permits. Only the information needed for the public disclosure of a vulnerability is excluded.
- Your personal data is not passed to third parties without your explicit consent.
- No legal action. If you act in accordance with this policy, we will not seek criminal prosecution against you. This does not apply if criminal intent was or is recognisably pursued.
- No NDA. We never ask you to sign a non-disclosure agreement.
- A trusted contact. We stay available to you throughout the whole process.
- Credit. If you ask, we name you (or your alias) and a link of your choice on our acknowledgements page once the process is complete. We don't run a bug bounty programme at this time.
4. Response times
Unless a report was submitted anonymously:
- a person (not an automated system) answers a report, or an update to one, within five working days;
- detailed feedback after analysis follows within ten working days. It either confirms or rejects the vulnerability, asks the questions we need answered to understand it, or explains why the analysis takes longer, with a commitment to another update within ten working days.
No report is closed by a single analyst: closing or rejecting a report always needs a second person to agree.
5. Disclosure
Validated and verified vulnerabilities are publicly disclosed within 90 days. If there is a valid reason for delaying a fix, this period can be extended once by another 90 days, in close consultation with the national CSIRT. Only the national CSIRT can grant further extensions. Disclosure takes place at least in the European Vulnerability Database (EUVD) maintained by ENISA, and we agree the timing and content with you.
We notify our national CSIRT, CERT-Bund at the BSI, without undue delay about actively exploited vulnerabilities in our products or infrastructure. We keep it informed of new information, mitigations and their schedules, and coordinate with it.
6. When the process ends
We consider the CVD process for a report complete when:
- the report turns out to be unfounded;
- a vulnerability in a service we operate is fixed and has been publicly disclosed;
- a vulnerability in a product is mitigated or fixed by a suitable patch and has been publicly disclosed;
- you have not answered our technical or content questions for at least 30 days, so the report can only be processed in part or not at all; or
- the vulnerability has been publicly disclosed and, in consultation with the national CSIRT, a mitigation or fix can no longer be expected.
Unless you reported anonymously, we tell you without undue delay when the process for your report has ended.
7. Code of conduct
We ask you:
- not to exploit a vulnerability beyond what is needed to demonstrate it, and to cause no damage;
- not to attack our systems, for example through social engineering, spam, denial of service or brute force;
- not to access, manipulate or modify the data or systems of third parties (for example other users' accounts);
- not to offer exploits or tools for the vulnerability to others, whether for a fee or free of charge;
- to treat everyone involved with respect. There is no room for discrimination, sexism or insults, including in the name or link you ask us to publish.
If these rules are not followed, we will not list you on the acknowledgements page. We will still handle your report as well as we can.