VulKill hosted MCP for authorized source-code and website security scans, account credits and evidence-based reports.
# VulKill hosted security scanning
Connect to https://app.vulkill.com/api/mcp using the user's own VulKill account. The scanning engine runs remotely. Consult the current tool catalog for supported checks, permissions, limits and credits.
Scan only assets the user owns or is authorized to test. Explain which source content or target will be sent and obtain the user's permission and any required authorization declaration. Do not put API keys or passwords in chat or configuration. Follow the service's live-site ownership verification process.
The catalog describes checks across 12 areas; actual coverage depends on the submitted scope and scan mode:
1. Authentication, sessions and access control.
2. Business logic, payments and race conditions.
3. Injection, command execution and unsafe data handling.
4. Network, SSRF and file-system security.
5. Secrets, credentials and cryptography in submitted content.
6. Dependencies, supply-chain security and software inventory.
7. AI, LLM and MCP agent security.
8. Web3 and smart-contract source, where supported.
9. Containers, cloud and infrastructure configuration.
10. Mobile and desktop application source and configuration, where supported.
11. AST, data-flow, symbolic analysis and binary hardening, where supported.
12. Authorized website and API checks, triage and remediation guidance, where available.
A local file path or private repository is not automatically available to the remote service: use only supported, user-authorized transfer or integration methods. Respect request and credit limits. Track scan status and distinguish completed, skipped and failed checks; do not claim findings, patches or coverage without evidence. Reports may contain limited code evidence and file locations. See https://vulkill.com/privacy/.