Bug Bounty Scoping and Rules of Engagement
What you'll learn: How to define what hackers can and cannot test in your bug bounty program, ensuring productive security research without legal or operational chaos.
Why Scoping Matters
Imagine opening your home to security consultants but forgetting to tell them the basement is off-limits because it contains critical medical equipment. Bug bounty scoping works the same way—you must clearly define which assets researchers can probe and what methods they can use, preventing accidental damage while maximizing security coverage.
Without proper scoping, well-intentioned researchers might trigger production outages, access regulated data, or perform actions that appear malicious to your security team.
Identifying In-Scope Assets
In-scope assets are the targets you want tested. These typically include:
- Specific domains and subdomains (e.g.,
app.company.com,api.company.com) - Mobile applications (with version numbers)
- API endpoints and services
- Web applications and user-facing portals
List them explicitly. Ambiguity leads to confusion and wasted effort on both sides.
Understanding Testing Boundaries
Define acceptable testing methods:
- Automated scanning limits (rate limiting considerations)
- Social engineering boundaries (usually prohibited)
- Physical security testing (typically out-of-scope)
- Denial-of-service restrictions (never allowed)
Clarify data handling: Can researchers create test accounts? May they use real payment methods? What data retention policies apply?
Recognizing Out-of-Scope Items
Explicitly exclude:
- Third-party services you don't control
- Legacy or deprecated systems
- Internal development environments
- Certain vulnerability classes (e.g., open redirects, rate limiting issues, or vulnerabilities requiring already-compromised credentials)
This prevents report noise and focuses researchers on meaningful vulnerabilities aligned with your risk tolerance.
Key Takeaway: Clear scoping and rules of engagement transform bug bounty programs from chaotic experiments into structured security partnerships—protecting both your organization and the researchers trying to help secure it.