Legal
Security and Vulnerability Disclosure
What protects your data, what we deliberately do not claim, and the terms under which you can go looking for holes in it.
1. Reporting a vulnerability
Write to hello@streetshape.org with “Security report” in the subject.
Tell us the affected URL or component, the steps to reproduce it, what you think the impact is, and the minimum evidence that shows it. Do not send somebody else’s personal data and do not post a working exploit publicly. If you need to send sensitive evidence, ask us for a secure way first.
We acknowledge a report within 2 working days and keep you reasonably updated. That is a promise about answering, not a promise to fix everything within a fixed time.
There is no standing paid bug bounty, and no reward is promised unless we agree one separately. The same address is published at /.well-known/security.txt.
2. What you may test
This policy authorizes limited, good-faith testing of StreetShape’s own application functions, using your own accounts and your own data, within these boundaries:
- Use the least testing needed to demonstrate the issue.
- Do not deliberately access, copy, alter or delete anybody else’s information. If you reach something unexpectedly, stop and tell us.
- No denial-of-service, no load testing, no intrusive automated scanning.
- No phishing, social engineering, physical attacks, credential stuffing, extortion, or leaving anything behind after testing.
- Do not test third-party infrastructure under this policy — the providers on the subprocessor page have their own programmes and their own permission to give.
- Ask us before doing anything outside these boundaries, or where the scope is unclear.
3. Safe harbour and coordinated disclosure
Our commitment
For research that follows this policy, we will not start legal action against you, and we will not report it as unauthorized activity to your employer or your internet provider, solely because you did that research.
This binds us and only us. It cannot authorize access to somebody else’s systems, and it cannot bind an authority or another rights holder. That is a limit of what any company can promise, and pretending otherwise would be the one dishonest sentence on a page about honesty.
We ask you to coordinate disclosure and normally allow 90 days from a sufficiently detailed report for us to investigate and fix. A different period can be agreed, taking account of active exploitation and of protecting users. We will discuss disclosing sooner where that is right, and we will not ask you to keep something secret indefinitely.
We will credit you publicly if you want that. A report is welcome even when we disagree about severity or cannot reproduce it — that may affect how we prioritize it, and it does not turn a report into misconduct.
4. What is in place
Safeguards proportionate to the service and its risks, meant to protect confidentiality, integrity and availability:
- A specialist authentication provider, with two-factor options. Passwords and authentication secrets never reach our own systems.
- Access control in the database itself, through row-level security policies, rather than only in application code.
- Role separation for staff, with administrative actions restricted by role and logged — including the ones that were refused.
- Support sessions that are read-only, expire after no more than thirty minutes, record who opened one and why, and email the account holder that it happened.
- Input and upload validation, with image metadata — including GPS tags — stripped on upload.
- Encryption in transit, and provider-managed encryption at rest.
- Security headers, same-origin protections, dependency auditing and rate limiting.
None of this guarantees that a vulnerability, a misconfiguration, a misuse of privileged access or a loss of data cannot happen. The controls reduce the chance and the blast radius. That is what controls do.
5. What we do not have
The honest list
- No ISO 27001 certification.
- No SOC 2 report.
- No completed independent penetration test.
- No tested and guaranteed recovery time, and no published uptime commitment unless one is expressly agreed with you.
Two more, which are open questions rather than settled positions and appear in the list at the top of this page: whether direct access to the production database is logged, and whether a backup restore has ever actually been run. An untested backup is a hypothesis.
Payment security
Full card credentials never reach our systems — Paddle handles them as merchant of record. Using a merchant of record is not a certification, and it does not by itself discharge every payment-security responsibility we might have. It removes the largest one.
6. If there is a breach
We assess a suspected personal-data breach promptly. Where the law requires it we notify the competent authority without undue delay and, where feasible, within 72 hours of becoming aware — explaining any delay. We notify the people affected without undue delay where the high-risk threshold and the other statutory conditions are met.
Where we are acting as somebody else’s processor, we notify that controller without undue delay and follow the incident procedure we agreed with them.
A notification says what happened, what the likely consequences are, what we are doing about it and who to contact, so far as we know at the time. We would rather send a second message with more detail than delay the first one until everything is certain.
7. Getting in touch
Security reports to hello@streetshape.org. Anything about personal data, including a breach affecting you, is covered by the privacy policy and the data protection and your rights page.