Decepticon

Entra Conditional Access bypass — discover policy gaps, exploit legacy-auth protocols (IMAP/POP/SMTP-AUTH/EWS), spoof device/platform/UA/location conditions, abuse service-principals + app-based auth excluded from CA, and break-glass account misuse.

Kubernetes RBAC privilege escalation paths — ClusterRole/Role enumeration via `kubectl auth can-i --list`, abuse of pods/exec, pods/portforward, secrets get, escalate verb, bind verb, impersonate verb, system:masters group abuse, ServiceAccount token theft and reuse.

Kubernetes pod escape to node — privileged container abuse, hostPath mount escape, hostPID/hostIPC, capability misuse (SYS_ADMIN, SYS_PTRACE), runC CVE chains. Pivots from RCE-in-pod to full node compromise.

Docker / containerd socket mounted into a container → host RCE. Common in CI runners, GitOps controllers (ArgoCD, Flux), and 'Docker-in-Docker' setups. Single-command escape via `docker run --rm --privileged -v /:/host alpine chroot /host`.

High-impact container-runtime CVE catalog — runC Leaky Vessels (CVE-2024-21626/-23651/-23652/-23653), CVE-2022-0185 (FUSE/legacy-fs), CVE-2019-5736 (runC binary replace), CRI-O Dirty COW analogs, Kubernetes API server CVE-2019-11247 (custom-resource RBAC bypass). Fingerprint → match → exploit.

Container / Kubernetes attack category — pod escape, RBAC abuse, runtime CVE exploitation, socket-mount escape. Routing skill: identify the surface (pod-internal RCE vs API-level vs build-pipeline), then load the matching sub-skill.

Azure Managed Identity abuse — IMDS at 169.254.169.254 from compromised VM / App Service / Function, token exchange for Graph/ARM/KeyVault, federated workload identity abuse, hybrid AAD Connect MSOL credential extraction.

AWS IAM privilege escalation via `iam:PassRole` chains — Lambda/Glue/Sagemaker/EC2/ECS PassRole to a higher-priv role, AssumeRole chains across accounts, sts:GetCallerIdentity recon, account hijack via legacy root-mfa-bypass.

Enumerate AWS IAM policies, detect privilege escalation paths per Rhino Security Labs canonical 21 primitives.

Cloud exploitation lane — AWS IAM privesc, S3 takeover, k8s RBAC abuse, Terraform state leaks, cloud metadata pivoting.

Hunt XML External Entity flaws in parsers and validate file read / SSRF impact with strict negative controls.

Hunt vector / embedding weaknesses (OWASP LLM08:2025) — adversarial inputs against the RAG / similarity layer that cause cross-tenant leak, embedding-inversion privacy loss, semantic confusion, and retriever-driven prompt injection.

Hunt LLM unbounded consumption (OWASP LLM10:2025) — denial-of-wallet and denial-of-service against LLM endpoints via unrestricted prompt size, runaway tool loops, expensive model selection, and unauthenticated fan-out.

Trust boundary mapping and startup sequence audit for developer tools, CLI apps, and plugin systems. Load when the target is a developer tool, CLI, IDE extension, or any application that loads config from the current directory.

YARA rule writing from behavioral observations and TI report analysis — sample-to-rule conversion, condition optimization, performance tuning, and retrohunting on VirusTotal and ANY.RUN. Covers YARA/YARA-X syntax, yarGen automated generation, and production rule deployment.

Automated IOC extraction from threat reports, logs, and unstructured text — parse hashes, IPs, domains, URLs, email addresses, and CVEs. Covers regex-based extraction, defanging/refanging, bulk hash lookup, IOC deduplication, YARA rule generation from IOCs, and STIX/TAXII formatting for sharing.

ANY.RUN Threat Intelligence Lookup workflow — query hashes, domains, IPs, and behavioral indicators against ANY.RUN's sandbox corpus. Covers TI Lookup query syntax, search operators, free tier constraints, result correlation with engagement findings, and integration with sandbox analysis.

Hunt LLM system-prompt leakage (OWASP LLM07:2025) — exfiltration of the privileged system prompt revealing internal rules, secrets baked in, tool inventory, and business logic that should not be client-visible.

Hunt LLM supply-chain compromise (OWASP LLM03:2025) — malicious or backdoored models, datasets, adapters, plugins, MCP servers, and tokenizer / framework dependencies that ship inside an AI-integrated product.

Hunt server-side template injection across Jinja2/Twig/Freemarker/Velocity/Handlebars and validate progression from expression injection to code execution.

Hunt Server-Side Request Forgery (CWE-918) through taint analysis from user-controlled URLs to HTTP client sinks. Covers cloud metadata pivoting, DNS rebinding, gopher smuggling, and the IMDSv1 → IAM role chain that turns SSRF into RCE.

Hunt SQL injection (CWE-89) via source-level taint tracking. Covers string concat, format-string, ORM raw queries, second-order injection, and NoSQL injection in MongoDB/DynamoDB.

Hunt LLM sensitive-information disclosure (OWASP LLM02:2025) — leakage of PII, secrets, internal source, model details, and other-tenant data through model outputs, training-data extraction, or retrieval-side joins.

Hunt ReDoS (CWE-1333, Catastrophic Backtracking) — identify regexes with nested quantifiers or overlapping alternation that cause super-linear matching time, trace tainted input paths to regex sinks, demonstrate timing PoC, and validate with response-time delta. Covers PCRE/RE2/V8/Python re engine differences. Triggers on: 'ReDoS', 'regex denial', 'catastrophic backtracking', 'redos', 'regex complexity', 'nested quantifiers', 'regex amplification', 'CWE-1333'.

Hunt JavaScript prototype pollution (CWE-1321) — the 2023-2026 meta-vulnerability that chains into RCE, auth bypass, and SSRF on most Node.js stacks.