DASTA-AI vs Safeguard DAST

DASTA-AI is a dynamic application security testing platform for first-party web, API, and LLM applications. Its scanner crawls endpoints and forms, probes them with non-destructive payloads, and verifies findings with proof of concept. The site lists more than 40 verified checks and three policy tiers. For AI red teaming, it tests chatbots, agents, and RAG apps against OWASP LLM Top 10 attack suites, with references to the NIST AI RMF Measure function and EU AI Act Article 55. Automated risk analysis produces executive summaries and Key Risk Indicators. DASTA-AI says findings are sanitized before AI analysis and credentials are not sent to third-party services. A CLI supports interactive login or a DASTA_API_TOKEN for unattended pipelines, and CI/CD integration is listed. It runs on the web and as an API. The Demo plan is free for one scan; Personal is $100 a month for 30 scans per month. Pro has 300 scans per month and Enterprise lists unlimited scans and BYOK, but neither plan has a listed price. Red-team scans count as one scan, with judge calls capped per organization.

Safeguard DAST dynamically tests running applications, reporting findings alongside how much discovered attack surface the scan reached. Its graph-aware crawler uses Safeguard SAST call graphs, route tables, and authentication middleware. Endpoint inventories can come from OpenAPI, GraphQL introspection, gRPC reflection, and live traffic samples. Targeted payloads cover SQL, OS command, SSRF, XSS, and deserialization sink classes, with parameter shapes informing generation. Authenticated flows include form authentication, OAuth 2.0 refresh, JWT bearer, session cookies, SAML, and mutual TLS. Listed scan modes are Fast, Deep, Progressive, and Reconnaissance; Reconnaissance sends no probes. Coverage Score explains reached surface and stopping reasons, while stated coverage includes the OWASP Top 10 and API Top 10. JavaScript, JSON, and source maps are analyzed for exposed credentials, routes, endpoints, cloud references, personal data, configuration, and comments. Secret samples are redacted, ownership verification gates requests, and scope, rate, and budget limits constrain them. It supports self-hosted and web use, hybrid deployment, CI integrations, pull-request and IDE workflows, and listed infrastructure analyses. Pricing is freemium: Free plan; paid from $9/mo.