Vulnerability Disclosure Policy.
Safe-harbour, SLAs, security.txt.
A complete VDP your legal team will sign and your security team can actually operate. Aligned to ISO 29147, CISA BOD 20-01, and RFC 9116 — with the safe-harbour clause language that researchers expect to see before they file a report.
A VDP both sides will trust.
- →12-section policy in editable Markdown plus a PDF reference render
- →Safe-harbour clause language used by mature programmes — counsel-friendly, researcher-trusted
- →Pre-built /.well-known/security.txt sample aligned to RFC 9116
- →Severity rubric, response SLA table, and remediation targets
- →Notes on adapting the policy from VDP to paid bug-bounty without rewriting governance
All twelve sections at a glance.
- 01
Intent statement
Plain-English commitment that the organisation welcomes security research, will respond to reports in good faith, and will not pursue legal action against researchers acting under this policy.
- 02
Scope — in
Domains, sub-domains, mobile apps, APIs, and infrastructure within scope. Pre-built as a table so legal and security can edit the rows without rewriting the policy.
- 03
Scope — out
Third-party services, partner-operated systems, deprecated assets, marketing micro-sites — and the specific test classes excluded (DoS, social engineering, physical, automated scanners at high rate).
- 04
Safe-harbour clause
Researcher protections modelled on CISA / Department of Justice guidance — no CFAA / DPDP / IT Act prosecution for good-faith research within scope. Limits clearly stated.
- 05
Reporting channel
Primary intake (security@ alias or platform), backup channel, PGP key reference, expected information per report (asset, reproduction steps, impact, evidence), language preferences.
- 06
Response SLAs
Acknowledgement within 3 business days, triage within 10 business days, remediation targets by severity (critical 30 days, high 60, medium 90), public-disclosure coordination window of 90 days.
- 07
Severity rubric
How reports are scored (CVSS v4 baseline plus business-context modifier), how disagreements are resolved, examples of how each severity tier maps to remediation priority.
- 08
Coordinated disclosure
Default 90-day disclosure window, conditions for extension, conditions for early disclosure (active exploitation), credit conventions, opt-out for anonymous reporters.
- 09
Recognition & rewards
Whether the programme is paid (bug-bounty) or recognition-only (VDP), the eligibility rules, the appeal route, the hall-of-fame conventions, anti-fraud controls.
- 10
Out-of-scope conduct
What the researcher must not do — exfiltrate real data beyond a proof, modify or destroy data, social-engineer staff, brute-force credentials, post findings publicly before disclosure window closes.
- 11
/.well-known/security.txt
RFC 9116 reference contents — Contact, Expires, Encryption, Policy URL, Hiring, Preferred-Languages — with worked example and signing guidance.
- 12
Governance & review
Owning team, escalation route to legal and exec sponsor, annual review cadence, change-history log, and the version-pinned canonical URL for the policy.
Pair this VDP with.
Service
VAPT
Internal pentest cadence that complements the inbound reports a VDP attracts.
Service
Compliance
A documented VDP is increasingly an audit expectation under SOC 2 CC7 and ISO 27001 A.16.
Blog
Bug Bounty vs Pentest
When a VDP is enough, when to upgrade to a paid programme, and when neither replaces a pentest.
Blog
Supply Chain Attacks 2026
Why third-party scope rules in the VDP are now the highest-leverage policy decision.
Standing up a VDP from scratch?
The template is the easy day. Triage capacity, severity rubric calibration, and disclosure-window discipline are where most VDPs wobble. A 30-minute scoping call is free.