42Crunch for IoT / OT

middleBrick is a self-service API security scanner designed for environments where installing agents is impractical, including many IoT and OT deployments. You submit a reachable URL and receive a risk score from A to F with prioritized findings within under a minute. The scanner uses read-only methods (GET and HEAD) plus text-only POST for LLM probes, avoiding any runtime instrumentation or code access.

Because it operates without agents or SDKs, it works across languages, frameworks, and cloud configurations common in operational technology. The approach is intentionally passive in its probe set, avoiding intrusive payloads that could disrupt fragile OT processes. The scanner validates security headers, cookie attributes, and HTTPS configurations while respecting the operational availability of the services under test.

middleBrick maps findings to OWASP API Top 10 (2023), supporting audit evidence for control validation across security frameworks. Detection covers authentication bypasses, JWT misconfigurations such as alg=none or missing claims, and security header compliance including WWW-Authenticate checks.

The scanner identifies BOLA and IDOR via sequential ID enumeration and active adjacent-ID probing, and detects BFLA and privilege escalation through admin endpoint probing and role/permission field leakage. Input validation checks include CORS wildcard usage with credentials, dangerous HTTP methods, and debug endpoints, while rate limiting and resource consumption are assessed by inspecting rate-limit headers and oversized responses.

Data exposure detection includes PII patterns such as email addresses, Luhn-validated card numbers, and context-aware SSNs, as well as API key formats for AWS, Stripe, GitHub, and Slack. Error and stack-trace leakage is flagged, along with missing encryption signals like HTTPS redirects, HSTS, and cookie flags.

SSRF checks target URL-accepting parameters and body fields, with passive indicators of internal IP detection and known cloud metadata endpoint patterns. The scanner also reviews inventory management issues, including missing versioning, legacy path patterns, and server fingerprinting that can aid reconnaissance in sensitive environments.

The scanner includes LLM / AI Security testing with 18 adversarial probes across three scan tiers: Quick, Standard, and Deep. These probes cover system prompt extraction, instruction override attempts, DAN and roleplay jailbreaks, data exfiltration techniques, cost exploitation, encoding bypasses, translation-embedded injection, few-shot poisoning, markdown injection, multi-turn manipulation, indirect prompt injection, token smuggling, tool abuse, nested instruction injection, and PII extraction.

These checks are designed to be non-intrusive and informative, helping teams understand how conversational interfaces might be abused without executing disruptive payloads. Results highlight risky prompts and model behaviors that could affect availability, integrity, or confidentiality of AI-assisted workflows.

middleBrick parses OpenAPI 3.0, 3.1, and Swagger 2.0 with recursive $ref resolution, cross-referencing spec definitions against runtime findings. This highlights undefined security schemes, sensitive fields, deprecated operations, and missing pagination that may indicate over-permissive designs.

Authenticated scanning (Starter tier and above) supports Bearer, API key, Basic auth, and Cookie credentials, gated by domain verification using DNS TXT records or HTTP well-known files. Only a curated allowlist of headers is forwarded, ensuring controlled credential usage. Continuous monitoring options on Pro and Enterprise tiers provide scheduled rescans, diff detection, email alerts, and integration with CI/CD pipelines.