Loading run…
Loading run…
gemini-3.5-flash-lite → ministral-14b-2512 · prompt v2 · retrieval embedding (gemini embeddings), top 5 · temperature 0 · 2026-10-27 11:15
Overall score
90.5
Correctness
4.70 / 5
Faithfulness
4.64 / 5
Invented answers
0%
Missed answers
3%
Expected sources cited
94%
| Category | direct | paraphrase | multi doc | partial | negative | unanswerable | trap | conflict | french |
|---|---|---|---|---|---|---|---|---|---|
| Score | 91n=15 | 97n=6 | 56n=5 | 98n=4 | 100n=5 | 93n=6 | 94n=5 | 100n=1 | 97n=3 |
Drafted answer
All customer data stored in databases, file storage, and backups is encrypted at rest using AES-256. Encryption keys are managed in AWS Key Management Service and are rotated automatically every year. Additionally, Enterprise customers can provide and control their own encryption key through AWS KMS.
Reference answer
Yes. All customer data (databases, file storage, backups) is encrypted at rest with AES-256, keys managed in AWS KMS and rotated automatically every year.
Expected: Security whitepaper › Encryption
Grader · correctness 5 · faithfulness 5 · completeness 3
The assistant's answer is **correct** in stating that customer data is encrypted at rest with AES-256 and keys managed in AWS KMS with annual rotation, aligning fully with the reference. However, the **completeness** is slightly reduced because the reference does not explicitly mention the Enterprise-specific BYOK (Bring Your Own Key) capability-this is an unsupported addition (though not incorrect in itself). The **faithfulness** is high, as all claims are supported by the cited excerpts, including the BYOK detail (though the reference does not mention it). The assistant did not hallucinate or introduce unsupported facts beyond the reference scope, but the extra detail about BYOK is not in the reference and thus not required for full completeness. The **correctness** is 5 because the core facts match the reference, and the BYOK detail, while not in the reference, is not contradicted by it (the reference does not address BYOK at all).
Drafted answer
All connections to Vantrel Plan use TLS 1.2 or higher, with TLS 1.3 preferred and negotiated whenever the client supports it.
Reference answer
TLS 1.2 is the minimum; TLS 1.3 is preferred. SSL, TLS 1.0 and 1.1 are disabled, and HSTS is enforced.
Expected: Security whitepaper › Encryption
Grader · correctness 5 · faithfulness 5 · completeness 4
The assistant's answer aligns closely with the reference answer, correctly stating that TLS 1.2 is the minimum supported version, TLS 1.3 is preferred, and older protocols (SSL, TLS 1.0, and TLS 1.1) are disabled. The assistant also accurately mentions HSTS enforcement, though it is not explicitly stated in the assistant's answer. However, the assistant's answer is fully supported by the cited excerpt, which explicitly covers all these points. The only minor omission is HSTS enforcement, which is implied in the reference but not explicitly stated in the assistant's answer. This omission does not significantly detract from correctness or completeness, as the core facts are accurate and well-supported by the documentation.
Drafted answer
Yes, Vantrel has been certified for ISO/IEC 27001:2022 since September 2023. The certificate covers the development, operation and support of Vantrel Plan.
Reference answer
Yes, ISO/IEC 27001:2022 certified since September 2023, covering development, operation and support of Vantrel Plan, with annual surveillance audits.
Expected: Security whitepaper › Certifications and audit reports
Grader · correctness 5 · faithfulness 5 · completeness 5
The assistant's answer is **correct** as it fully aligns with the reference answer, including the certification date (September 2023), the ISO/IEC 27001:2022 standard, and the scope (development, operation, and support of Vantrel Plan). The **faithfulness** is perfect since every claim is directly supported by the cited excerpt from the Security Whitepaper. The **completeness** is also complete, as all key points from the reference are addressed without omission or addition of irrelevant details. The assistant did not hallucinate or extrapolate beyond the provided documentation, and the answer is concise yet thorough.
Drafted answer
We undergo an annual SOC 2 Type II examination covering the Security, Availability and Confidentiality Trust Services Criteria. The SOC 2 Type II report can be shared with customers and prospects under a non-disclosure agreement on request to security@vantrel.example.
Reference answer
Yes. Vantrel has an annual SOC 2 Type II (Security, Availability, Confidentiality; period Jan 1 - Dec 31, report available around March). It is shared under NDA on request to security@vantrel.example.
Expected: Security whitepaper › Certifications and audit reports
Grader · correctness 5 · faithfulness 5 · completeness 5
The assistant's answer is factually consistent with the reference answer, correctly stating that Vantrel undergoes an annual SOC 2 Type II examination covering Security, Availability, and Confidentiality, and that the report is shared under NDA upon request. The cited excerpts fully support these claims, including the audit period (Jan 1 - Dec 31) and the sharing policy. The assistant's answer is also complete, covering all key points from the reference (availability, scope, and sharing conditions). No contradictions or omissions are present, and the answer aligns perfectly with the documentation provided.
Drafted answer
For EU-hosted tenants, customer data is stored and processed in the European Union, with the primary location in AWS eu-west-3 (Paris, France) and the disaster recovery location in AWS eu-central-1 (Frankfurt, Germany).
Reference answer
On AWS in eu-west-3 (Paris, France), with disaster recovery in eu-central-1 (Frankfurt, Germany). Data at rest, including backups, stays in the selected region.
Expected: Infrastructure hosting › Hosting regions
Grader · correctness 3 · faithfulness 5 · completeness 3
The assistant's answer correctly identifies the primary and disaster recovery locations for EU-hosted customer data (AWS eu-west-3 and eu-central-1, respectively) and aligns with the reference answer. However, the assistant omits the critical detail about **data at rest, including backups, staying in the selected region**, which is explicitly stated in the reference and the Infrastructure and Hosting excerpt. The assistant's answer is supported by the cited excerpts, but the omission of this key point affects completeness. The reasoning and correctness are otherwise accurate, with no wrong statements or unsupported claims.
Drafted answer
In the event of a major incident affecting an entire AWS region for Vantrel Plan, our Recovery Point Objective is 1 hour and our Recovery Time Objective is 4 hours.
Reference answer
RPO of 1 hour and RTO of 4 hours.
Expected: Business continuity › Objectives
Grader · correctness 5 · faithfulness 5 · completeness 5
The assistant's answer directly matches the reference answer in both RPO (1 hour) and RTO (4 hours), demonstrating full alignment with the key facts. The cited excerpts explicitly state these objectives, providing complete support for the claims made. The assistant also correctly specifies the scope (major incident affecting an entire AWS region for Vantrel Plan), which aligns with the reference context. No additional unsupported details or omissions are present.
Drafted answer
Databases are backed up continuously with point-in-time recovery available for the last 7 days, and full encrypted snapshots are taken daily and retained for 35 days. Backups containing customer data will eventually expire according to this backup retention schedule after contract termination.
Reference answer
Continuous backups with 7-day point-in-time recovery, plus daily encrypted full snapshots retained for 35 days, replicated to the DR location. Restores are tested monthly.
Expected: Business continuity › Backups
Grader · correctness 3 · faithfulness 3 · completeness 3
The assistant's answer correctly captures the core backup frequency (continuous with 7-day point-in-time recovery and daily encrypted snapshots retained for 35 days) and aligns with the reference answer. However, the assistant adds an unsupported claim about the expiration of backups *after contract termination* (the reference does not specify this timing, and the cited excerpts only mention the 35-day retention schedule without linking it to contract termination). The assistant also omits the key detail about **monthly restore testing** (present in the excerpts but not in the reference answer). The reference answer includes the DR replication and restore testing, which are partially addressed in the excerpts but not fully reflected in the assistant's response. The assistant's claim about backups expiring after contract termination is not explicitly supported by the cited excerpts, making it an unsupported extrapolation. The reference answer does not contradict the assistant's core points but is more precise in scope and includes additional critical details (e.g., DR replication, restore testing).
Drafted answer
We support SAML 2.0 and OpenID Connect (OIDC) Single Sign-On protocols on Business and Enterprise plans.
Reference answer
SAML 2.0 and OpenID Connect, on Business and Enterprise plans. Tested with Microsoft Entra ID, Okta and Google Workspace. Enforcing SSO disables password login.
Expected: Identity access › Customer authentication
Grader · correctness 4 · faithfulness 5 · completeness 3
The assistant's answer correctly identifies the supported SSO protocols (SAML 2.0 and OpenID Connect) and the applicable plans (Business and Enterprise) as stated in the reference answer and the cited excerpt. However, the assistant's response omits key details present in the reference answer, such as the tested identity providers (Microsoft Entra ID, Okta, and Google Workspace) and the enforcement behavior (disabling password login when SSO is enforced). These omissions affect completeness. The assistant's answer is fully supported by the cited excerpt, so faithfulness is not an issue. Correctness is also high since the main points align with the reference, though the missing details slightly reduce the score for completeness.
Drafted answer
Yes, user provisioning and deprovisioning through SCIM 2.0 is available on the Enterprise plan.
Reference answer
Yes, SCIM 2.0 is supported, on the Enterprise plan only.
Expected: Identity access › User provisioning
Grader · correctness 5 · faithfulness 5 · completeness 5
The assistant's answer aligns perfectly with the reference answer, confirming SCIM 2.0 support exclusively on the Enterprise plan. The cited excerpt directly supports the claim, and no additional unsupported details or omissions are present. The assistant correctly avoids inferring support on non-Enterprise plans, as the reference and excerpt clarify manual management applies elsewhere. Completeness is achieved by addressing the core requirement (SCIM 2.0 support) and its plan restriction without deviation.
Drafted answer
For Enterprise customers, we commit to a monthly uptime of 99.95%, excluding scheduled maintenance.
Reference answer
99.95% monthly uptime for Enterprise (excluding scheduled maintenance), with service credits of 10% or 25% of the monthly fee.
Expected: Sla support › Availability commitment
Grader · correctness 5 · faithfulness 5 · completeness 3
The assistant's answer correctly states the **99.95% monthly uptime commitment for Enterprise customers**, which matches the reference answer. However, the reference also explicitly mentions **service credits of 10% or 25% of the monthly fee** in case of downtime, which the assistant omitted. The cited excerpts support the uptime percentage but do not mention service credits, so the assistant's answer is **partially incomplete** and **lacks a key detail** from the reference. The assistant did not hallucinate or misrepresent facts, but the omission of service credits reduces the completeness score.
Drafted answer
We notify the customer of any personal data breach affecting the customer's data without undue delay, and in any case within 48 hours after becoming aware of it.
Reference answer
Without undue delay and in any case within 48 hours after becoming aware of it.
Expected: Data processing agreement › Personal data breach notification
Grader · correctness 5 · faithfulness 5 · completeness 5
The assistant's answer correctly aligns with the **reference answer** by stating the notification timeframe as 'without undue delay and in any case within 48 hours.' The cited excerpts fully support this claim, as the **Data Processing Agreement (summary)** explicitly mentions the same timeframe. The assistant's answer is concise, covering the key point without unnecessary details, and does not introduce any unsupported claims or omissions. The **Incident Response** excerpt, while relevant to broader incident communication, does not contradict or alter the core requirement for breach notification timelines. Thus, the answer is fully correct, faithful, and complete.
Drafted answer
We test Vantrel Plan using an independent third-party security firm at least once a year and additionally before major architectural changes. The most recent penetration test was completed in March 2026 and identified no critical findings and two high-severity findings, both of which were remediated within 30 days and verified by the testing firm during a retest.
Reference answer
At least annually by an independent third party, plus before major architectural changes. The last test (March 2026) found no critical and two high findings, both remediated within 30 days and verified by retest.
Expected: Vulnerability management › Penetration testing
Grader · correctness 5 · faithfulness 5 · completeness 5
The assistant's answer aligns perfectly with the reference answer in terms of frequency (annual + before major architectural changes), the timing of the last test (March 2026), and the results (no critical findings, two high-severity findings remediated within 30 days). The cited excerpts fully support all claims, including the remediation verification by the testing firm. The assistant did not introduce any unsupported details or omissions, and the answer is concise yet complete. The reference does not mention any contradictions or additional key points beyond what the assistant covered.
Drafted answer
We must remediate critical vulnerabilities within 7 days. Exceptions must be documented, include compensating controls, and be approved by the CISO.
Reference answer
Critical vulnerabilities must be remediated within 7 days (high: 30 days, medium: 90 days).
Expected: Vulnerability management › Vulnerability management
Grader · correctness 5 · faithfulness 5 · completeness 3
The assistant's answer correctly states the 7-day remediation timeline for critical vulnerabilities, which matches the reference answer. However, the assistant omits the additional timelines for high (30 days) and medium (90 days) vulnerabilities, which are key points in the reference. The cited excerpts fully support the assistant's claims, including the exception process. The omission of the other severity timelines reduces completeness, but the core fact (critical = 7 days) is correct and supported. The exception process is also correctly cited and aligns with the reference, though it is not explicitly mentioned in the reference answer (only implied by the assistant's inclusion of the full excerpt).
Drafted answer
Our sub-processors include Amazon Web Services EMEA SARL for cloud hosting, storage, and backups, Cloudflare, Inc. for CDN, Web Application Firewall, and DDoS protection, and Datadog, Inc. for infrastructure and application monitoring and logs. We also use Stripe Payments Europe, Ltd. for billing and subscription payments, Zendesk, Inc. for customer support ticketing, SendGrid for transactional emails, and Microsoft Ireland Operations Ltd. for AI features.
Reference answer
AWS (hosting), Cloudflare (CDN/WAF/DDoS), Datadog (monitoring), Stripe (billing), Zendesk (support), SendGrid/Twilio (transactional emails), Microsoft Azure OpenAI Service (AI features).
Expected: Subprocessors › Infrastructure sub-processors · Subprocessors › Service sub-processors
Grader · correctness 5 · faithfulness 5 · completeness 5
The assistant's answer aligns closely with the reference answer in terms of the sub-processors listed, though it includes specific legal entity names (e.g., *Amazon Web Services EMEA SARL* instead of just *AWS*) and clarifies some purposes (e.g., *transactional emails* explicitly includes *password resets*). The assistant’s answer is fully supported by the cited excerpts, which provide granular details on each sub-processor’s role and data location. The only minor deviation is the inclusion of *Twilio Inc.* in parentheses for SendGrid, which is not explicitly stated in the reference but is implied (SendGrid is owned by Twilio). The reference does not mention legal entity names, but the core sub-processors and their purposes match perfectly, ensuring correctness and completeness. The assistant’s answer is exhaustive and faithful to the documentation, with no unsupported claims or omissions of key points.
Drafted answer
Customer data is never used to train or fine-tune AI models, neither by Vantrel nor by its AI sub-processor.
Reference answer
No. Customer data is never used to train or fine-tune AI models, neither by Vantrel nor by its AI sub-processor.
Expected: Ai features policy › Use of customer data
Grader · correctness 5 · faithfulness 5 · completeness 5
The assistant's answer is factually consistent with the reference answer, which states that customer data is never used to train or fine-tune AI models by Vantrel or its sub-processor. The cited excerpts directly support this claim, and the assistant does not introduce any unsupported additions or omissions. The answer is concise and aligns perfectly with the reference, covering all key points without unnecessary details or inaccuracies.
Drafted answer
We support Single Sign-On via SAML 2.0 and OpenID Connect on Business and Enterprise plans, and Okta is one of the tested identity providers.
Reference answer
Yes, via SAML 2.0 or OIDC SSO on Business and Enterprise plans; Okta is a tested identity provider.
Expected: Identity access › Customer authentication
Grader · correctness 5 · faithfulness 5 · completeness 5
The assistant's answer correctly states that SSO via SAML 2.0 and OIDC is supported on Business and Enterprise plans, and that Okta is a tested identity provider, aligning perfectly with the reference answer. The cited excerpts fully support these claims, and the assistant does not introduce any unsupported details or contradictions. The answer is concise and covers all key points from the reference without omissions or additions beyond the scope of the question or documentation. The reference explicitly confirms the facts, and the assistant’s response is both faithful and complete.
Drafted answer
In the event of a major incident affecting an entire AWS region, such as the Paris primary location, our Recovery Point Objective is 1 hour.
Reference answer
At most about 1 hour of data (RPO 1 hour), with service restored from the Frankfurt DR region within 4 hours (RTO).
Expected: Business continuity › Objectives · Infrastructure hosting › Hosting regions
Grader · correctness 5 · faithfulness 5 · completeness 3
The assistant's answer correctly identifies the **Recovery Point Objective (RPO) of 1 hour** as stated in the reference and supported by the cited excerpts. However, it omits the **Recovery Time Objective (RTO) of 4 hours**, which is a key point in the reference answer. While the assistant's response is factually correct for the RPO, it fails to address the full scope of the question by excluding the RTO detail. The cited excerpts clearly mention both RPO and RTO, but the assistant only references the RPO. The assistant did not hallucinate or introduce unsupported claims, but the omission of the RTO reduces completeness. The answer is also not misleading, as it does not contradict the reference or documentation.
Drafted answer
Vantrel employees do not have standing access to customer data. Support engineers can only access a customer workspace when the customer grants temporary access from the administration console.
Reference answer
No standing access. Support can only access a workspace when the customer grants temporary access (max 72 hours), which is logged and visible in the audit trail.
Expected: Identity access › Vantrel employee access to customer data
Grader · correctness 5 · faithfulness 5 · completeness 3
The assistant's answer correctly states that Vantrel employees lack standing access to customer data and that support engineers require temporary access granted by the customer. However, the assistant's answer omits the **time limit (72 hours)** and the **audit trail visibility** mentioned in both the reference and the cited excerpts, which are key details. While the core point (no standing access) is correct, the omission of these qualifications reduces the score for completeness. The assistant's claims are fully supported by the cited excerpts, so faithfulness is high. The reference answer also includes the time limit and audit trail, which the assistant missed, further justifying the lower completeness score.
Drafted answer
Enterprise customers can provide and control their own encryption key through AWS KMS, and revoking the key makes the customer's data unreadable by Vantrel.
Reference answer
Yes, Enterprise customers can use customer-managed keys (BYOK) via AWS KMS; revoking the key makes the data unreadable by Vantrel.
Expected: Security whitepaper › Encryption
Grader · correctness 5 · faithfulness 5 · completeness 5
The assistant's answer aligns closely with the reference answer, correctly stating that Enterprise customers can use customer-managed keys (BYOK) via AWS KMS and that revoking the key renders the data unreadable by Vantrel. The cited excerpts directly support these claims, particularly the **Customer-managed keys (BYOK)** section. The assistant's phrasing is accurate and does not introduce any unsupported additions or omissions of key points. The reference answer and assistant's answer are functionally identical in substance, with no contradictions or missing qualifications (e.g., no mention of plan restrictions or additional caveats).
Drafted answer
We perform background checks on all new employees before their start date, where permitted by local law.
Reference answer
Yes, background checks are performed on all new employees before their start date, where permitted by local law.
Expected: Security whitepaper › Employee security
Grader · correctness 5 · faithfulness 5 · completeness 5
The assistant's answer directly matches the reference answer in both content and phrasing, confirming that background checks are performed on all new employees before their start date, where permitted by local law. The cited excerpt fully supports this claim, and no additional unsupported statements or omissions are present. The assistant did not introduce any incorrect facts or extrapolate beyond the documentation. Since the reference answer is fully aligned with the assistant's response, all criteria are met at the highest level.
Drafted answer
For Business plan customers, support is available during business hours from Monday to Friday, with a first-response target of 4 hours for P1 service unavailability issues. For Enterprise plan customers, support is available 24/7 for P1 issues, with a first-response target of 1 hour. Starter customers have access to email support on a best-effort basis.
Reference answer
Enterprise: P1 first response within 1 hour, 24/7. Business: support is business hours only, P1 first response within 4 hours.
Expected: Sla support › Support
Grader · correctness 5 · faithfulness 5 · completeness 5
The assistant's answer aligns perfectly with the reference answer for both Business and Enterprise plans, correctly stating the first-response targets (4 hours for Business during business hours, 1 hour for Enterprise 24/7). The assistant also included an accurate but non-key detail about Starter customers (best-effort email support), which does not detract from correctness or completeness. All cited statements are directly supported by the provided excerpts, and no key points from the reference are missing. The only minor deviation is the inclusion of Starter plan details, which are not part of the reference but are supported by the documentation and do not affect the core evaluation criteria.
Drafted answer
Yes, some of our sub-processors may process data outside the EU. Specifically, Stripe Payments Europe, Ltd. processes billing and subscription payments in the EU and US, SendGrid handles transactional emails in the US, and Amazon Web Services EMEA SARL provides cloud hosting, storage, and backups in regions selected by the customer, which can include the US.
Reference answer
For EU tenants, hosting (AWS), monitoring (Datadog), support (Zendesk) and AI (Azure EU Data Zone) stay in the EU. Cloudflare handles traffic in transit on its global network, Stripe operates in EU/US, and SendGrid processes email addresses and notification content in the US. Transfers are covered by SCCs; Vantrel Inc. is under the EU-U.S. Data Privacy Framework.
Expected: Subprocessors › Infrastructure sub-processors · Subprocessors › Service sub-processors · Data processing agreement › International data transfers
Grader · correctness 3 · faithfulness 5 · completeness 3
The assistant's answer correctly identifies that some sub-processors (Stripe, SendGrid, and AWS) process data outside the EU, aligning with the reference answer. However, the assistant omits key details from the reference, such as the EU-only processing for Zendesk, Datadog (for EU tenants), and Azure EU Data Zone, as well as the mention of Cloudflare's global network handling traffic in transit. The assistant also incorrectly states that AWS *may* process data outside the EU (implying uncertainty), whereas the reference clarifies that AWS's data location is *selected by the customer* (EU or US). The cited excerpts support the assistant's claims about Stripe, SendGrid, and AWS, but the assistant's phrasing about AWS's data location is misleadingly vague. The reference also includes critical context about SCCs and the EU-U.S. Data Privacy Framework, which the assistant omits entirely. The assistant's answer is factually correct in identifying the *which* sub-processors operate outside the EU but lacks precision and completeness in explaining *how* or *why* this occurs, as well as the safeguards in place (e.g., SCCs).
Drafted answer
Service availability incidents are published on the public status page at status.vantrel.example, with updates at least every 60 minutes for SEV1 incidents. In addition, customers affected by a security incident involving their data are notified by email by our Security team, in accordance with the timelines defined in the Data Processing Agreement.
Reference answer
Availability incidents on the status page (updates at least every 60 min for SEV1); customers whose data is affected are emailed by Security within the DPA timeline (48 hours); a root cause analysis is shared within 10 business days after a SEV1.
Expected: Incident response › Customer communication · Data processing agreement › Personal data breach notification
Grader · correctness 5 · faithfulness 5 · completeness 3
The assistant's answer aligns closely with the reference answer for the key points regarding customer communication during security incidents. It correctly mentions the status page updates for SEV1 incidents and the email notifications for affected customers, both of which are supported by the cited excerpts. However, the assistant omitted the specific timeline of **48 hours** for email notifications (as per the DPA) and the **10 business days** for sharing the root cause analysis (RCA) after a SEV1 incident. While the assistant's answer is mostly faithful to the excerpts, it misses these critical details, which are explicitly stated in the reference answer. The omission of these timelines reduces the completeness score, though correctness and faithfulness remain high for the included facts.
Drafted answer
Customer users can enable multi-factor authentication using a TOTP authenticator app or a FIDO2/WebAuthn security key, and administrators can make MFA mandatory for all users of their workspace, while SMS-based MFA is not supported. The documentation does not describe the MFA options for Vantrel staff accessing production.
Reference answer
Customers: TOTP or FIDO2/WebAuthn, can be made mandatory by admins, no SMS. Staff: production admin access goes through a bastion with SSO, MFA and session recording; direct SSH disabled.
Expected: Identity access › Customer authentication · Infrastructure hosting › Network security
Grader · correctness 3 · faithfulness 3 · completeness 3
The assistant's answer correctly describes the MFA options for customer users (TOTP, FIDO2/WebAuthn, and mandatory enforcement by admins) and accurately states that SMS-based MFA is not supported, all of which align with the reference and cited excerpts. However, the assistant claims the documentation does not describe MFA options for Vantrel staff accessing production, which is incorrect-the reference explicitly states that staff production access requires a bastion with SSO, MFA, and session recording, with direct SSH disabled. This omission of key details for staff access results in a lower score for completeness. Faithfulness is high since all customer-related claims are supported by the excerpts, but the unsupported claim about staff documentation hurts the score. The main point (customer MFA options) is correct, but the assistant missed critical staff-specific details, lowering correctness and completeness.
Drafted answer
Infrastructure and application logs are centralized in Datadog, utilizing the EU site for EU-hosted tenants and the US site for US-hosted tenants. Infrastructure logs are retained for 12 months, and audit logs are retained for 13 months. However, the documentation does not specify the retention period for application logs.
Reference answer
Infrastructure and application logs are centralized in Datadog (EU site for EU tenants), retained 12 months. Customer audit logs are retained 13 months and exportable via API or streamed to a SIEM (Enterprise).
Expected: Infrastructure hosting › Logging and monitoring · Identity access › Audit logs
Grader · correctness 1 · faithfulness 1 · completeness 3
The assistant's answer contains a **correctness** issue by stating that application logs are retained for an unspecified period (claiming 'documentation does not specify'), while the reference answer explicitly states they are retained for **12 months** (same as infrastructure logs). The assistant also **hallucinates** the US site usage for EU-hosted tenants (the reference and excerpts clarify only EU site for EU tenants). The **faithfulness** score drops due to unsupported claims (e.g., US site for EU tenants) and incorrect omission of retention for application logs. **Completeness** is partially met, as key points (Datadog centralization, 12-month retention for infrastructure, 13-month for audit logs, and SIEM streaming) are covered, but critical details (e.g., application log retention) are missing or misrepresented. The assistant’s citation of excerpts is accurate but misinterpreted (e.g., conflating EU/US site usage).
Drafted answer
When Vantrel Assist is enabled, we send only the data required to answer a specific request to Microsoft Ireland Operations Ltd. (Azure OpenAI Service), which is our AI sub-processor. For EU-hosted tenants, this service is deployed in the EU Data Zone.
Reference answer
Only if a workspace Admin enables Vantrel Assist (off by default). Then only the data needed for the request is sent to Microsoft Azure OpenAI Service in the EU Data Zone; it is never used for training.
Expected: Ai features policy › Vantrel Assist · Ai features policy › Use of customer data · Subprocessors › Service sub-processors
Grader · correctness 3 · faithfulness 5 · completeness 3
The assistant's answer correctly identifies that data is only sent to Microsoft Ireland Operations Ltd. (Azure OpenAI Service) when Vantrel Assist is enabled, and that only the data required for the request is sent. However, the assistant's answer omits the critical detail that Vantrel Assist is **off by default** and must be enabled by a workspace Admin, which is explicitly stated in the reference answer. Additionally, the assistant's answer does not explicitly state that the data is **never used for training**, which is a key point in the reference and excerpts. The cited excerpts fully support the assistant's claims about data usage and the sub-processor, but the omission of the 'off by default' detail and the training restriction reduces correctness and completeness scores. Faithfulness is high since all stated claims are supported by the excerpts, but the answer is not fully complete or correct due to these omissions.
Drafted answer
Vantrel runs a private bug bounty program with invited researchers. The reward amounts are not documented.
Reference answer
Yes, a private bug bounty program with invited researchers. Reward amounts are not documented.
Expected: Incident response › Reporting a vulnerability
Grader · correctness 5 · faithfulness 5 · completeness 5
The assistant's answer correctly identifies the existence of a private bug bounty program with invited researchers, which aligns perfectly with the reference answer and the cited excerpt. However, the assistant's claim that 'reward amounts are not documented' is accurate according to the reference answer, but the assistant did not explicitly state that the reference answer itself confirms this lack of documentation (it only says 'reward amounts are not documented'). The assistant's answer is faithful to the excerpts and covers the key points of the reference, though it does not add any unsupported details. The reference answer explicitly states that reward amounts are not documented, so the assistant's answer is correct in this regard as well.
Drafted answer
The requested information about the specific firm that performed our last penetration test is not available in the documentation.
Reference answer
An independent third-party security firm, in March 2026. The firm's name is not documented.
Expected: Vulnerability management › Penetration testing
Grader · correctness 5 · faithfulness 5 · completeness 5
The assistant correctly identifies that the information about the specific firm performing the last penetration test is not documented, which aligns perfectly with the reference answer. However, the assistant did not cite any excerpts to support this claim, which is a minor oversight since the reference itself confirms the lack of documentation. The answer is concise, accurate, and covers the key point of the reference without adding or omitting details.
Drafted answer
Enterprise customers can stream audit logs to their SIEM via webhook.
Reference answer
Enterprise customers can stream audit logs to their SIEM via webhook; specific supported SIEM products are not listed.
Expected: Identity access › Audit logs
Grader · correctness 5 · faithfulness 5 · completeness 5
The assistant's answer correctly states that enterprise customers can stream audit logs to their SIEM via webhook, which aligns perfectly with the reference answer. The reference explicitly mentions that no specific SIEM products are listed, and the assistant does not claim otherwise, avoiding incorrect assumptions. The cited excerpt directly supports the assistant's statement, and the answer covers the key point of the reference without omissions or unsupported claims. Since the reference confirms the information is not documented beyond the general capability, the assistant's response is accurate and complete.
Drafted answer
Our Data Protection Officer can be contacted at dpo@vantrel.example.
Reference answer
The DPO can be contacted at dpo@vantrel.example; the DPO's name is not documented.
Expected: Company overview › Security and compliance organization
Grader · correctness 5 · faithfulness 5 · completeness 3
The assistant's answer correctly identifies the contact email for the Data Protection Officer (DPO) as `dpo@vantrel.example`, which matches the reference answer and is explicitly stated in the cited excerpt. However, the assistant omits the key point from the reference answer that the DPO's name is **not documented**, which is a critical detail for completeness. The assistant’s answer is factually correct but incomplete. The cited excerpt fully supports the email address provided, so faithfulness is high. The omission of the 'name not documented' detail reduces the completeness score significantly, as it is a key point in the reference answer.
Drafted answer
Custom roles are not currently supported in Vantrel Plan.
Reference answer
No, custom roles are not currently supported. Four default roles exist (Admin, Modeler, Contributor, Viewer) plus dimension-member-level restrictions.
Expected: Identity access › Roles and permissions
Grader · correctness 5 · faithfulness 5 · completeness 5
The assistant's answer is factually correct and aligns perfectly with the reference answer, which states that custom roles are not supported and only four default roles exist. The assistant's response is directly supported by the cited excerpt, which explicitly mentions 'Custom roles are not currently supported.' The answer is concise and covers the key point without omissions or unsupported claims. Since the reference and assistant's answer both confirm the lack of custom roles, there is no room for misinterpretation or missing qualifications.
Drafted answer
Vantrel Plan is delivered exclusively as a multi-tenant Software-as-a-Service application accessed through a web browser. We do not offer on-premise, private cloud or self-hosted deployments.
Reference answer
No. Vantrel Plan is SaaS only; no on-premise, private cloud or self-hosted deployment.
Expected: Company overview › Product and delivery model
Grader · correctness 5 · faithfulness 5 · completeness 5
The assistant's answer is fully aligned with the reference answer, accurately stating that Vantrel Plan is a SaaS-only solution with no on-premise, private cloud, or self-hosted deployment options. The cited excerpt directly supports this claim, and the assistant does not introduce any unsupported details or omissions. The answer is concise, covering all key points without unnecessary elaboration.
Drafted answer
SMS-based MFA is not supported.
Reference answer
No, SMS-based MFA is not supported (TOTP and FIDO2/WebAuthn are).
Expected: Identity access › Customer authentication
Grader · correctness 5 · faithfulness 5 · completeness 5
The assistant's answer ('SMS-based MFA is not supported') is factually correct and aligns perfectly with the reference answer, which explicitly states that SMS-based MFA is not supported while listing TOTP and FIDO2/WebAuthn as supported alternatives. The assistant's answer is directly supported by the cited excerpt, which explicitly mentions 'SMS-based MFA is not supported' under the MFA section. The answer is also complete, as it covers the key point of the reference without omissions or unnecessary details. No contradictions or unsupported claims are present.
Drafted answer
No, data cannot be hosted in Australia as no other hosting regions are currently offered beyond the EU and US regions.
Reference answer
No. Only EU (Paris/Frankfurt) and US (N. Virginia/Oregon) regions are offered.
Expected: Infrastructure hosting › Hosting regions
Grader · correctness 5 · faithfulness 5 · completeness 5
The assistant's answer correctly aligns with the reference answer, stating that data cannot be hosted in Australia and only EU and US regions are offered. The cited excerpts explicitly list the available regions (EU and US) and confirm no other regions are currently available, directly supporting the assistant's claim. The answer is concise, avoids unsupported extrapolations, and covers the key point of regional restrictions without omissions or inaccuracies.
Drafted answer
We are working toward conformance with WCAG 2.1 level AA, but an accessibility conformance report (VPAT) is not yet available.
Reference answer
Not yet. Vantrel is working toward WCAG 2.1 AA, but a VPAT is not available yet.
Expected: Product faq › Is Vantrel Plan accessible to users with disabilities?
Grader · correctness 5 · faithfulness 5 · completeness 5
The assistant's answer is factually consistent with the reference answer, correctly stating that Vantrel is working toward WCAG 2.1 AA and that a VPAT is not yet available. The cited excerpt directly supports both claims, ensuring faithfulness. The answer also covers all key points from the reference without omissions or unsupported additions, meeting completeness requirements.
Drafted answer
The requested information regarding FedRAMP authorization is not in the documentation.
Reference answer
Not in the documentation.
Grader · correctness 5 · faithfulness 5 · completeness 5
The assistant correctly identifies that the information regarding FedRAMP authorization for Vantrel Plan is not documented, which aligns perfectly with the reference answer. Since no excerpts were cited, faithfulness is irrelevant (defaulting to 5 for not claiming unsupported facts), and completeness is fully satisfied by the accurate reflection of the reference. The answer is concise and avoids any incorrect assumptions or extrapolations.
Drafted answer
The provided documentation does not contain information on whether we sign a HIPAA Business Associate Agreement.
Reference answer
Not in the documentation.
Grader · correctness 5 · faithfulness 3 · completeness 5
The assistant correctly identifies that the documentation does not address the question of signing a HIPAA Business Associate Agreement (BAA), aligning perfectly with the reference answer. However, the assistant did not cite any excerpts to support this claim, which is unusual since the reference explicitly states the information is *not* documented. While the answer is factually correct, the lack of citation (even if the documentation is silent) slightly weakens the justification. The completeness and correctness are fully satisfied, but faithfulness is penalized due to the absence of cited evidence.
Drafted answer
The provided documentation does not contain information about the coverage amount of our cyber insurance policy.
Reference answer
Not in the documentation.
Grader · correctness 5 · faithfulness 3 · completeness 5
The assistant correctly identifies that the coverage amount of the cyber insurance policy is not documented, which aligns perfectly with the reference answer. However, the assistant did not cite any excerpts to support this claim, which is unusual since the reference answer explicitly states the information is not documented. Since the reference confirms the absence of documentation, the assistant’s answer is factually correct, but the lack of cited excerpts slightly detracts from faithfulness. Completeness is fully satisfied as the answer directly addresses the question without omissions or additions.
Drafted answer
The requested information about a source code escrow agreement is not documented in the available excerpts.
Reference answer
Not in the documentation.
Grader · correctness 5 · faithfulness 5 · completeness 5
The assistant correctly identifies that the information about a source code escrow agreement is not documented, which aligns perfectly with the reference answer. Since no excerpts were cited, faithfulness is irrelevant (defaulting to 5 for not claiming unsupported facts), and the answer is concise yet complete. The assistant did not hallucinate or infer anything beyond the reference's guidance, and no key points were missed or added incorrectly.
Drafted answer
The provided documentation does not contain information regarding IP allowlisting to restrict user access.
Reference answer
Not in the documentation.
Grader · correctness 5 · faithfulness 5 · completeness 5
The assistant correctly identifies that the documentation does not address IP allowlisting for restricting user access, which aligns perfectly with the reference answer. Since the reference explicitly states the information is not documented, the assistant’s response is accurate and complete. However, the assistant did not cite any excerpts, which is irrelevant in this case because the reference confirms the absence of documentation, but it would have been better practice to explicitly state this (e.g., 'No excerpts were found').
Drafted answer
The provided documentation does not contain information regarding the use of post-quantum cryptography.
Reference answer
Not in the documentation.
Grader · correctness 5 · faithfulness 3 · completeness 5
The assistant correctly identifies that the documentation does not address the use of post-quantum cryptography, aligning perfectly with the reference answer. However, the assistant did not cite any excerpts, which is a minor procedural issue but does not affect the correctness or completeness of the response. Since the reference explicitly states the information is not documented, the answer is fully accurate and complete, though the lack of cited excerpts prevents a perfect faithfulness score (as no excerpts were provided to verify).
Drafted answer
The requested information is not in the documentation.
Reference answer
Not in the documentation. Only Vantrel's SOC 2 Type II is documented; the SOC 1 mentioned is AWS's, not Vantrel's.
Grader · correctness 5 · faithfulness 5 · completeness 5
The assistant correctly identifies that the SOC 1 report is not documented, which aligns perfectly with the reference answer. However, the assistant did not cite any excerpts, which is a minor oversight since the reference explicitly states the information is not documented (and thus no citation is needed). The answer is concise, accurate, and covers the key point of the reference. Since no excerpts were cited, faithfulness cannot be evaluated negatively, but the absence of citations is technically a deviation from the schema's expectation (though harmless in this case).
Drafted answer
The provided documentation does not contain information about the number of people who work in our security team.
Reference answer
Not in the documentation. Only the total headcount (~240) and the CISO/Security Committee are mentioned.
Grader · correctness 5 · faithfulness 5 · completeness 5
The assistant correctly identifies that the documentation does not provide the specific number of people in the security team, aligning perfectly with the reference answer. However, the assistant did not cite any excerpts to support this conclusion, which is a minor oversight since the reference itself confirms the absence of such information. The answer is concise, accurate, and covers the key point of the reference (lack of documentation).
Drafted answer
The provided documentation does not contain information about the mean time to resolve P1 incidents.
Reference answer
Not in the documentation. Only first-response targets are documented, explicitly not resolution times.
Grader · correctness 5 · faithfulness 3 · completeness 5
The assistant correctly identifies that the information about the mean time to resolve P1 incidents is not documented, which aligns perfectly with the reference answer. However, the assistant did not cite any excerpts to support this claim, which is a minor but notable omission since the reference answer explicitly states that only first-response targets are documented (not resolution times). The answer is concise and accurate in its core assertion, but the lack of citation weakens its faithfulness score slightly. Completeness is fully satisfied since the reference answer only requires stating that the information is not documented.
Drafted answer
The requested information is not available in the documentation.
Reference answer
Not in the documentation. The documented certifications are ISO 27001 and SOC 2 Type II.
Grader · correctness 5 · faithfulness 3 · completeness 5
The assistant correctly identifies that the information about CSA STAR certification is not documented, which aligns perfectly with the reference answer. However, the assistant did not cite any excerpts to support this claim, which is unusual given the task requirements. Since the reference explicitly states the absence of this information, the assistant’s answer is factually correct and complete, but the lack of citation prevents a full score for faithfulness. The reference also provides additional context (ISO 27001 and SOC 2 Type II certifications), which the assistant did not include, though this does not affect correctness or completeness for this specific question.
Drafted answer
The provided documentation does not contain information about an emergency phone number to call during a security incident.
Reference answer
Not in the documentation. Only email (security@vantrel.example) and the status page are documented.
Grader · correctness 5 · faithfulness 5 · completeness 5
The assistant correctly identifies that the documentation does not provide an emergency phone number, aligning perfectly with the reference answer. However, the assistant did not cite any excerpts to support this claim, which is a minor but notable omission since the reference explicitly states the absence of such information in the documentation. The answer is concise and avoids incorrect assumptions or unsupported claims, focusing solely on the documented communication channels (email and status page).
Drafted answer
Following contract termination, customers have 30 days to export their data. According to the Security Whitepaper, customer data is permanently deleted from production systems within 90 days of contract termination, while the Data Processing Agreement states it is deleted within 30 days after the end of the 30-day grace period.
Reference answer
The documentation is inconsistent: the Security Whitepaper says within 90 days of termination, while the DPA says within 30 days after a 30-day export grace period (i.e. ~60 days). Both sources should be cited and the inconsistency flagged.
Expected: Security whitepaper › Data retention and deletion · Data processing agreement › Return and deletion of data
Grader · correctness 5 · faithfulness 5 · completeness 5
The assistant correctly identified the conflicting timelines (90 days in the Security Whitepaper vs. ~60 days in the DPA) and cited the exact excerpts supporting both versions. However, the assistant did not explicitly flag the inconsistency as strongly as the reference answer (e.g., by stating 'The documentation is inconsistent' upfront). The assistant’s answer is factually accurate, fully supported by the cited excerpts, and covers all key points from the reference, though it could have emphasized the contradiction more clearly. The minor omission of the legal retention clause in the DPA does not detract from correctness or completeness, as it is a secondary detail not required for answering the core question.
Drafted answer
Pour les clients de la région EU (par défaut), l'emplacement principal d'hébergement est AWS eu-west-3 à Paris en France, et les données de nos clients au repos y restent.
Reference answer
Oui pour la région EU (par défaut) : AWS eu-west-3 à Paris, avec reprise d'activité à Francfort (Allemagne). Une région US existe aussi.
Expected: Infrastructure hosting › Hosting regions
Grader · correctness 5 · faithfulness 5 · completeness 3
The assistant's answer correctly identifies the primary hosting location for the EU region as AWS eu-west-3 (Paris, France) and confirms that customer data at rest remains within the selected region, which aligns with the reference answer and the cited excerpts. However, the assistant omits the mention of the US region entirely, which is a key point in the reference answer. Additionally, the assistant does not explicitly mention the disaster recovery location (Frankfurt, Germany), though this is a minor omission since the question focuses on the primary hosting location. The cited excerpts fully support the assistant's claims about the EU region, but the omission of the US region and disaster recovery location reduces completeness.
Drafted answer
Oui, nous soutenons l'authentification unique (SSO) via SAML 2.0 et OpenID Connect sur les plans Business et Enterprise, et Microsoft Entra ID fait partie des fournisseurs d'identité testés.
Reference answer
Oui, SSO SAML 2.0 et OIDC sur les plans Business et Enterprise ; Microsoft Entra ID fait partie des fournisseurs testés.
Expected: Identity access › Customer authentication
Grader · correctness 5 · faithfulness 5 · completeness 5
The assistant's answer is factually consistent with the reference answer, correctly stating that SSO via SAML 2.0 and OIDC is supported on Business and Enterprise plans and that Microsoft Entra ID is a tested identity provider. The cited excerpts directly support all claims made in the assistant's answer, including the specific protocols (SAML 2.0 and OIDC) and the inclusion of Microsoft Entra ID among tested providers. The answer is also complete, covering all key points from the reference without omissions or unsupported additions. The assistant did not introduce any incorrect or misleading details, and the reasoning aligns perfectly with the documentation provided.
Drafted answer
The requested information is not in the documentation.
Reference answer
Pas dans la documentation.
Grader · correctness 5 · faithfulness 5 · completeness 5
The assistant correctly identifies that the information is not documented, which aligns perfectly with the reference answer. Since no excerpts were cited, faithfulness is irrelevant (defaulting to 5 as per instructions). The assistant’s response is concise and fully covers the reference’s key point (absence of documentation), without adding or omitting anything. No contradictions or hallucinations are present.