Decepticon

Hunt LLM prompt injection and tool-call hijacking in modern AI-integrated applications (CWE-1427). Covers indirect injection via RAG, tool abuse, exfiltration chains, and jailbreak-to-RCE pivots on agentic systems.

Systematic pattern exhaustion methodology. Load after finding any confirmed vulnerability to search for all instances of the same root cause pattern across the codebase.

Hunt directory traversal and archive traversal (ZipSlip/TarSlip) from user input to filesystem operations.

Extract a functional clone of a black-box ML model via prediction API queries, and infer whether specific records were in the training set (membership inference).

Hunt LLM misinformation / overreliance (OWASP LLM09:2025) — confident-but-wrong outputs that flow into downstream automated decisions, compliance reports, customer communications, or autonomous code commits without verification.

Hunt improper LLM output handling (OWASP LLM05:2025) — downstream code that trusts unstructured model output and renders / executes / shells it without sanitisation, producing XSS, SSRF, SQL injection, RCE, and SSTI via the model channel.

Hunt Insecure Direct Object Reference (CWE-639) — missing authorization checks on object IDs. Covers horizontal vs vertical privilege escalation, UUID vs integer guessing, and GraphQL introspection-driven IDOR discovery.

Hunt LLM excessive agency (OWASP LLM06:2025) — agentic systems granted too many tools, too broad permissions per tool, or unsupervised authority to act on the user / business behalf, producing financial loss, data loss, or destructive operations from a single bad token.

Hunt insecure deserialization (CWE-502) across Python pickle, Java ObjectInputStream / Jackson / SnakeYAML, .NET BinaryFormatter / DataContractJson, PHP unserialize, Ruby Marshal/YAML.load, and Node.js vm. Direct path to unauthenticated RCE.

Hunt LLM training-data and model poisoning (OWASP LLM04:2025) — adversarial inputs that bias future model behaviour through fine-tuning, RLHF, or continuous-learning loops.

Hunt OS command injection (CWE-78) — user input reaching shell, exec, or system calls. Covers argument-array bypasses, path confusion, and template-string injection in modern frameworks.

Build chains from XSS into account takeover or privileged action execution.

Build and validate SSRF pivot chains toward metadata/infra control and final code execution impact.

Build chains where IDOR enables privilege escalation and high-impact control-plane actions.

Build chains where leaked or weak credentials pivot across services to privileged access.

Bug bounty white-box hunting methodology. Load when the target is an open-source project with a security advisory program, bug bounty, or responsible disclosure policy.

Hunt authentication/authorization bypass in route guards, role checks, tenant boundaries, and state-machine transitions.

Craft adversarial examples that cause trained ML classifiers to misclassify at inference time — image recognition, malware detectors, IDS, spam filters.

Root pointer for the analyst's vulnerability research playbooks. Load this first at iteration start to see the full catalog of vuln-class and chain-building skills.

NTLM relay deep-dive — `ntlmrelayx` configuration matrix (SMB, LDAP, LDAPS, HTTP, RPC, IMAP, MSSQL), SMB-signing bypass, target selection (DC for DCSync, ADCS for cert, LAPS reader for cleartext), session relay vs cracking trade-off, multi-relay (forward auth from one victim to many).

NetExec (CrackMapExec successor) — unified SMB/LDAP/MSSQL/WinRM/RDP/SSH/FTP/VNC protocol auth + post-auth modules. 200+ modules incl. BloodHound auto-ingest, ESC1-15 scanning, PrintNightmare, LDAP relay.

Extract LAPS-managed local administrator passwords from AD computer objects (ms-Mcs-AdmPwd / msLAPS-Password).

Request Kerberos TGS tickets for SPN-bound service accounts and crack offline with hashcat — classic AD priv-esc primitive.

Abuse replication rights (DS-Replication-Get-Changes + GetChangesAll) to dump krbtgt and arbitrary user NT hashes from a DC.

Authentication coercion against Windows / AD — PetitPotam (MS-EFSR), PrinterBug (MS-RPRN), DFSCoerce (MS-DFSNM), ShadowCoerce (MS-FSRVP), Coercer.py meta-tool. Force a Windows machine to NTLM-authenticate to attacker, then relay or crack offline.