Trust & Security

How ASE protects your data

ASE is built on dedicated single-tenant infrastructure, data isolation by design, and deterministic security controls. This page documents our architecture, data handling practices, and compliance posture for security teams evaluating ASE.

01 Security Architecture

ASE operates on a dedicated-instance model designed for isolation and client data control. Reports are automatically purged from the server after download or after the retention window expires (default: 72 hours).

Dedicated Instance Architecture

Single-tenant isolation with client-side data custody

Each client is provisioned a dedicated server instance with its own database, encryption keys, and static IP address
No infrastructure is shared between clients: isolation is enforced at the infrastructure level
A lightweight local agent runs on the client's own machine, providing a web interface and offline-capable local storage
Assessment results are downloaded and stored locally: the client retains the authoritative copy of all reports
Server-side data is encrypted at rest (AES-256-GCM) and in transit (TLS 1.3)
Reports are automatically purged from the server after download or after the retention window expires (default: 72 hours)

Isolation guarantee

Every client is assigned a dedicated compute instance with a persistent static IP address, enabling WAF whitelisting and consistent network identity. Instances are never shared between clients. When an assessment completes, working data (logs, agent state, temporary evidence) is purged immediately. The encrypted report is retained until the client downloads it or the retention window expires (default: 72 hours). This isolation is enforced at the infrastructure level, not by application logic.

02 Data Protection

ASE handles credentials, tokens, and sensitive target data that appear during security assessments. Our data protection controls are designed to meet the requirements of regulated industries.

Automatic secret redaction

Credentials, API keys, tokens, and PII discovered during assessments are automatically redacted from logs and agent context. Only hashed references are retained for finding verification.

Zero cross-contamination

Assessment data (evidence, credentials, agent memory, vector stores) is ephemeral and cannot leak between clients. Each engagement starts from a clean state on the client's dedicated instance.

Encryption at rest and in transit

All data is encrypted at rest using AES-256-GCM and in transit using TLS 1.3. Key management follows industry best practices with regular rotation.

Retention and destruction

Once delivered, the report is yours. ASE does not retain copies: reports are automatically purged from our servers after download or after the retention window expires (default: 72 hours). You are responsible for storing and managing your own reports. No data is used for model training. Destruction is cryptographic and verified.

No model training on client data

Client assessment data, targets, findings, and reports are never used to train, fine-tune, or improve ASE's AI models. This is a contractual commitment, not just a policy.

Data processing agreement

ASE provides a Data Processing Agreement (DPA) to all clients, covering data categories, processing purposes, sub-processors, and breach notification obligations.

03 AI Safety Controls

Autonomous AI agents require deterministic safety controls at the platform level. ASE enforces scope, limits destructive actions, and provides full observability.

Deterministic scope enforcement. Target scope (URLs, domains, IP ranges) is enforced at the network and platform level via allowlists. The AI agent cannot override or expand the scope through prompt manipulation. Scope restrictions are not prompt-based.

Safe mode by default. ASE operates in non-destructive mode by default. No data modification, no data deletion, no account lockout, no production exploitation. All vulnerabilities are confirmed through safe, non-impactful validation techniques that pose zero risk to the target environment.

Emergency stop (kill switch). Every assessment can be stopped immediately by the user. Automatic halt triggers on: time limit exceeded, iteration limit exceeded, unexpected target response, or scope boundary hit.

Data sanitization. Credentials, API keys, tokens, and personally identifiable information discovered during assessments are automatically redacted from internal storage. Sensitive information is never persisted in plain text within ASE infrastructure.

Prompt injection defenses. The agent treats all target output (HTML, API responses, banners, error messages) as untrusted input. Output filtering, input sanitization, and sandboxed execution prevent the target from manipulating agent behavior.

Egress control. Assessment instances have restricted outbound connectivity. Only traffic to the defined target scope and required ASE services is permitted. All other egress is blocked at the network level.

Independent Safety Judge. A dedicated safety component, architecturally separate from the execution agents, evaluates every action before it is permitted. This separation ensures that security controls cannot be bypassed through prompt injection, agent manipulation, or emergent behavior. The Safety Judge operates on deterministic rules, not model reasoning.

Deterministic command controls. Every command an agent attempts to execute is intercepted and evaluated against deterministic allowlists and blocklists before execution. Arguments are analyzed for risk, and operations such as file deletion, credential modification, or database writes are blocked at the platform level, independent of the AI model's reasoning.

Automatic target health monitoring. ASE continuously monitors the target's availability and response patterns during assessments. If the target shows elevated error rates, increased latency, or becomes unresponsive, testing is automatically paused until conditions recover. Agents that repeatedly attempt blocked actions are also detected and halted.

AI-powered vulnerability validation. Findings are verified through an independent validation layer that executes controlled probes to confirm exploitability with observable, measurable evidence. This process drastically reduces false positives, ensuring reported vulnerabilities are reproducible and actionable.

04 Infrastructure Security

ASE's cloud infrastructure is managed with security-first defaults. Each client operates on dedicated, isolated infrastructure with no shared resources.

Dedicated client instances

Each client is assigned a dedicated compute instance with a static IP address. Instances are single-tenant and isolated. Assessment data is purged after client download or after the retention window expires (default: 72 hours). The instance is retained exclusively for that client to ensure IP consistency for WAF whitelisting.

Hosted on SOC 2 Type II certified infrastructure

Our infrastructure runs on enterprise cloud providers that hold SOC 2 Type II certification, covering security, availability, and confidentiality trust services criteria.

Least privilege IAM

No static access keys. Instance roles follow least-privilege with scoped, short-lived credentials. IMDSv2 enforced. No IAM role has access to other clients' resources.

Monitoring and detection

Threat detection and runtime monitoring enabled across all assessment infrastructure. Anomalous behavior triggers automated alerts and instance isolation.

Network isolation

Each instance runs in an isolated network segment. No lateral movement between assessment instances. Ingress limited to the ASE control plane; egress limited to the target scope.

Fixed and predictable pricing

ASE charges a fixed fee per URL assessment. There are no GPU costs, no usage-based scaling, and no risk of runaway billing from agent loops. Clients know the exact cost before an assessment begins.

05 Privacy Policy

This section describes what data ASE collects, how it is processed, and how long it is retained.

Data we collect

Category	Data	Purpose	Retention
Account	Name, email, company	Authentication, billing, support	Duration of account + 30 days
Target	URLs, domains, IPs provided by client	Scope definition for assessment	Duration of assessment only
Assessment	Findings, evidence, PoCs, logs	Report generation	Encrypted at rest (AES-256-GCM) until client download, then purged. Retention window: 72 hours (default). Client retains the authoritative copy on their own infrastructure
Credentials	Tokens, keys found during testing	Finding documentation	Redacted in logs, hashed references only
Usage	Assessment count, execution time	Billing, platform health	Aggregated, anonymized after 90 days

What we do not do

We do not sell, share, or license client data to third parties
We do not use client data to train, fine-tune, or evaluate AI models
We do not retain assessment data beyond the client's chosen retention window
We do not access client reports without explicit authorization
We do not store credentials or tokens in plaintext at any stage

Sub-processors

ASE maintains a list of sub-processors involved in service delivery. This list is updated before any change, with 30 days' notice provided to affected clients.

Your rights

Access: request a copy of all data ASE holds about you
Deletion: request immediate deletion of your account and all associated data
Portability: export your data in a portable format
Correction: update your account information at any time
Objection: opt out of anonymized usage analytics

06 Compliance & Standards

ASE aligns its security practices with recognized industry standards and frameworks.

Standard	Status	Scope
OWASP Top 10	Covered	Full coverage in assessment methodology
CVSS 4.0	Adopted	All findings scored using CVSS 4.0
CWE	Adopted	All vulnerabilities classified by CWE ID
PTES	Aligned	Assessment methodology follows PTES phases
ISO/IEC 42001	Aligned	AI management system practices
GDPR	Aligned	Data processing, retention, and rights management

Secure development lifecycle

Internal code review and static analysis on every release
Vetted open-source tooling: ASE integrates industry-standard security tools (nmap, Nuclei, ffuf, sqlmap, among others) with pinned versions, integrity verification, and isolated execution environments to manage supply chain risk
Regular third-party penetration testing of the ASE platform itself
Responsible disclosure program for security researchers
Change management with staged rollouts and rollback capability

07 Incident Response

ASE maintains an incident response plan covering security incidents, data breaches, and service disruptions.

Detection. Automated monitoring across all infrastructure layers. Anomaly detection on assessment behavior, network traffic, and authentication events.

Containment. Affected instances are isolated automatically. Assessment data is preserved for forensic analysis in a quarantined environment.

Notification. Affected clients are notified within 72 hours of a confirmed breach. Notification includes scope, impact, timeline, and remediation steps.

Remediation. Root cause analysis, corrective action, and post-incident report provided to affected clients. Lessons learned are integrated into platform controls.

To report a security vulnerability in the ASE platform, contact security@ase.run. We acknowledge reports within 48 hours and provide an initial assessment within 5 business days.

08 Responsible Use & AI Governance

ASE is designed to be used by authorized security professionals against targets they have explicit permission to test.

Platform controls

Assessments require authenticated users with verified identity
Target scope must be defined before any assessment can begin
Each assessment is tied to an authenticated user account
Abuse detection monitors for unauthorized or malicious usage patterns
Accounts found violating acceptable use policy are suspended immediately

For your contracts and RoE

We recommend that clients using ASE update their Rules of Engagement (RoE) and Statements of Work (SOW) to explicitly mention the use of AI-powered autonomous security testing tools. ASE can provide template language for this purpose upon request.

AI governance

Internal AI usage policy governing model selection, evaluation, and deployment
Regular benchmarking of assessment quality against manual testing baselines
Human validation recommended for all critical and high severity findings
Transparent about AI limitations: ASE complements, does not replace, expert review

Questions about our security posture?

Security teams evaluating ASE can review our Security Whitepaper, Data Processing Agreement, and Sub-processor List directly. For security questionnaire responses (SIG, CAIQ) and penetration test executive summary under NDA, contact security@ase.run or your ASE account representative.