- Top reward
- No cash rewards
- Top severity paid
- Critical
- Assets in scope
- 5
- Public CVEs
- 0
- Last change
At a glance
What to know before you start testing.
- No cash rewards
- A disclosure program: you report for credit, not money.
- Report on HackerOne
- Submit through the program page.
- Partial safe harbor
- Some legal protection; read the terms.
- Publish after a fix
- Coordinate the timing with the company.
Rewards
What HackerOne lists as the payout for each severity.
This program does not pay cash rewards.
- Safe harbor
-
Partial
While the program has Safe harbour policy, it's important to avoid any business impact on user, data, availability related to your testing.
Summarised with AI from the program policy on HackerOne, read 12 Nov 2026, and checked word for word against it. Read the original
Scope
5 assets you may test, 1 excluded.
Web 5
- api.lovable.dev Critical No bounty
- lovable.dev Critical No bounty
- lovable.net Critical No bounty
- mcp.lovable.dev Critical No bounty
- oauth.lovable.dev Critical No bounty
Out of scope (1)
- lovable.app
Rules
What the policy asks of you, and the bug types it won't reward, quoted from Lovable VDP (via HackerOne)'s policy. "We" in these lines means the program, not this site.
While testing
-
Always provide detailed reports with reproducible steps. E.g python script demonstrating the exploit.
-
Avoid automated scanning, DAST, fuzzing lovable API and endpoints.
-
Social engineering (e.g., phishing, vishing, smishing) is strictly prohibited.
-
Only interact with accounts you own or with the explicit permission of the account holder.
-
Never target real users or companies.
-
Consult with Lovable staff before performing any destructive or sensitive actions.
-
Submit one vulnerability per report unless you need to chain vulnerabilities to provide impact.
-
When duplicates occur, we only triage the first report received (provided that it can be fully reproduced).
-
Always add session header to identify as legitimate researcher.
Not accepted
-
TLS, infrastructure configuration, security headers, clickjacking issues unless can be exploited.
-
AI or tooling prompt leakage.
-
Unsafe AI instructions.
-
AI jailbreaking, prompt injection, other types of AI attacks without proven impact.
-
Injecting malicious code, vulnerabilities, or files into websites built with Lovable
-
Input validation issues unless part of a successful exploit chain.
-
Information disclosure unless part of a successful exploit chain.
-
Vulnerable dependencies or components unless part of a successful exploit chain.
Summarised with AI from the program policy on HackerOne, read 12 Nov 2026, and checked word for word against it. Read the original
Changes
Tracked since 11 Nov 2026. Last checked 4 minutes ago.
No changes yet. When the rewards or scope move, you will see it here and in the RSS feed.
Before we tracked it
Scope and reward changes recorded before this page started tracking it.
-
Scope added
lovable.net
-
Rules updated
Added: “Lovable Linux Desktop App”
-
Scope added
mcp.lovable.dev
-
Scope removed
auth.lovable.dev
-
Scope added
api.lovable.dev
-
Scope added
oauth.lovable.dev
-
Scope added
lovable.app (out of scope)
-
Scope added
lovable.dev
Public CVEs
Published vulnerabilities in this vendor's products, from the NVD.
No CVEs matched yet. We match NVD records by vendor name, so products sold under another name can be missing.
Original policy
Copied from hackerone.com on 12 Nov 2026. The summaries on this page are made from it; the live page may have changed since.
Disclosure Policy
- Do not discuss any vulnerabilities (even resolved ones) outside of the program without express consent from Lovable.
- Follow HackerOne's disclosure guidelines.
Program Rules
- Always provide detailed reports with reproducible steps. E.g python script demonstrating the exploit.
- Avoid automated scanning, DAST, fuzzing lovable API and endpoints.
- Social engineering (e.g., phishing, vishing, smishing) is strictly prohibited.
- Only interact with accounts you own or with the explicit permission of the account holder.
- Never target real users or companies. Consult with Lovable staff before performing any destructive or sensitive actions. While the program has Safe harbour policy, it's important to avoid any business impact on user, data, availability related to your testing.
- Submit one vulnerability per report unless you need to chain vulnerabilities to provide impact. Multiple vulnerabilities caused by one underlying issue will be treated as one valid report.
- When duplicates occur, we only triage the first report received (provided that it can be fully reproduced).
- Always add session header to identify as legitimate researcher.
Identifying yourself as a researcher
Researchers should add the following header to all HTTP requests:
X-HackerOne-Research: [H1 username]
While you're welcome to create multiple accounts for research purposes, we ask that you use your HackerOne email alias when doing so. You can find instructions on how to use your email alias here: https://docs.hackerone.com/en/articles/8404308-hacker-email-alias
Focus of the Program
- Lovable Editor (lovable.dev) application security
- Generic project takeover via prompt injection (e.g. injections delivered via Lovable Cloud DB)
- Development sandbox escape or takeover
- Cross-tenant data access
- Coding agent escape (get access to underlying APIs or resources)
- Broken access controls or privilege escalation within workspace
Out-of-scope
(This does not mean we do not care about the problems below, but we want to focus VDP on novel exploits)
- TLS, infrastructure configuration, security headers, clickjacking issues unless can be exploited.
- AI or tooling prompt leakage.
- Unsafe AI instructions.
- AI jailbreaking, prompt injection, other types of AI attacks without proven impact.
- Injecting malicious code, vulnerabilities, or files into websites built with Lovable is out of scope. As creators fully control their own sites, self-reflected vulnerabilities (such as adding alert(1)) do not represent valid security issues in the Lovable platform. Reports must demonstrate a clear impact on other users or the Lovable infrastructure to be considered in scope. Findings limited to a reporter’s own website or any customer-built apps (*.lovable.app, except if created by Lovable) will be marked as Not Applicable (N/A).
- Input validation issues unless part of a successful exploit chain.
- Information disclosure unless part of a successful exploit chain.
- Vulnerable dependencies or components unless part of a successful exploit chain.
- Lovable Linux Desktop App
We appreciate your time and effort and thank you for helping keep Lovable and our users safe!