No server-side request forgery or outside internal network access
Customers must not use the Service to conduct server-side request forgery (SSRF), or to access internal network resources outside the Customer's authorized environment. This applies through any Service component, with the HTTP Request node, webhook endpoints and workflow import named as examples.
Supporting quotations
Security and infrastructure
Conduct server-side request forgery (SSRF) or access internal network resources outside Customer's authorized environment via any Service component (HTTP Request node, webhook endpoints, workflow import, or otherwise)
Qualifications & exceptions
Internal network access is prohibited only when it is outside the Customer's authorized environment.
Covers any Service component, including the HTTP Request node, webhook endpoints and workflow import.
Not stated here
The sentence does not say whether "outside Customer's authorized environment" also limits the SSRF prohibition, or what counts as the authorized environment.
Security & Safeguards · Restriction or exception
No exfiltrating credentials or secrets
Customers must not use the Service to exfiltrate credentials, tokens, API keys or secrets that are stored in or accessible through the Service.
Supporting quotations
Exfiltrate credentials, tokens, API keys, or secrets stored in or accessible via the Service
Qualifications & exceptions
Covers secrets stored in or accessible via the Service.
Not stated here
The policy does not say whose credentials are covered or how this differs from a customer moving its own credentials.
Security & Safeguards · Restriction or exception
No hiding malicious traffic or abusing computing resources
Customers must not use the Service to obscure or launder the origin of malicious network traffic, including through proxies, tunnels or relays. They also must not abuse computational or network resources, with cryptocurrency mining as an example, including through sandbox or code execution environments.
Supporting quotations
Use the Service to obscure or launder the origin of malicious network traffic, including through proxies, tunnels, or relays
abuse computational or network resources (e.g. cryptocurrency mining), including through sandbox or code execution environments
Qualifications & exceptions
The traffic prohibition concerns malicious network traffic.
Resource abuse includes sandbox or code execution environments.
Not stated here
The policy does not define "abuse" of computational or network resources or set any usage threshold.
Security & Safeguards · Restriction or exception
No attempts to gain unauthorised access to third-party systems
Customers must not use the Service to attempt to gain unauthorised access to third-party systems, accounts or credentials. Brute-force attacks, credential stuffing and automated enumeration are named as examples.
Supporting quotations
Attempt to gain unauthorised access to third-party systems, accounts, or credentials, including via brute-force, credential stuffing, or automated enumeration
Qualifications & exceptions
Covers third-party systems, accounts or credentials.
Security & Safeguards · Restriction or exception
No denial-of-service attacks
Customers must not use the Service to launch denial-of-service attacks, whether against the Service or against third-party systems.
Supporting quotations
Launch denial-of-service attacks against the Service or third-party systems
Qualifications & exceptions
Covers attacks against the Service and against third-party systems.
Security & Safeguards · Restriction or exception
No spyware or covert monitoring without valid consent
Customers must not use the Service to deploy spyware or covert monitoring capabilities without valid consent.
Supporting quotations
Deploy spyware or covert monitoring capabilities without valid consent
Qualifications & exceptions
The prohibition applies when valid consent is absent.
Not stated here
The policy does not say whose consent counts as valid or what makes it valid.
Security & Safeguards · Restriction or exception
No jailbreaking or circumventing AI model safety filters
Customers must not use the Service to jailbreak, prompt-inject or otherwise circumvent AI model safety filters that are accessible through the Service.
Supporting quotations
Jailbreak, prompt-inject, or otherwise circumvent AI model safety filters accessible via the Service
Qualifications & exceptions
Covers AI model safety filters accessible via the Service.
Security & Safeguards · Restriction or exception
No hosting or distributing malicious content
Customers must not use the Service for malicious hosting. That means hosting, executing, storing or distributing malware, phishing websites, credential-harvesting services, or other malicious code, content or infrastructure through the Service, including its sandbox environments.
Supporting quotations
Malicious hosting, such as hosting, executing, storing, or distributing malware, phishing websites, credential-harvesting services, or other malicious code, content, or infrastructure through the Service, including its sandbox environments.
Customers must secure credentials and promptly report suspected misuse
Customers must maintain reasonable access controls over credentials and secrets in the Service. They must also promptly notify n8n of any suspected misuse or security incident.
Supporting quotations
Enforcement and Security
Customer must maintain reasonable access controls over credentials and secrets in the Service, promptly notify n8n of any suspected misuse or security incident
Qualifications & exceptions
Access controls must be reasonable.
Notice must be prompt and covers suspected misuse or a security incident.
Not stated here
The policy does not say how to notify n8n or what counts as reasonable access controls.
n8n may suspend or remove abusive activity on its hosted infrastructure without notice
Where n8n reasonably believes that n8n-hosted infrastructure, including sandbox environments, is being used in violation of this AUP or poses a security risk, it may immediately and without prior notice suspend or terminate the affected processes, environments or network connections. It may also block or remove associated malicious content. It says it will give notice where reasonably practicable. This sits alongside the prohibition on malicious hosting, which includes the Service's sandbox environments.
Supporting quotations
Where n8n reasonably believes that any n8n-hosted infrastructure, including sandbox environments, is being used in violation of this AUP or poses a security risk, n8n may immediately, without prior notice, suspend or terminate affected processes, environments or network connections, and block or remove associated malicious content.
n8n will provide notice where reasonably practicable.
including its sandbox environments
Qualifications & exceptions
Applies to n8n-hosted infrastructure, including sandbox environments.
Requires a reasonable belief of violation or security risk.
Action reaches affected processes, environments or network connections and associated malicious content.
Notice is given where reasonably practicable.
Not stated here
The policy does not say whether the notice comes before or after the action, or how soon.