Results for “adversarial-attacks”

Benchmarks large audio-language models against adversarial audio attacks using the CAA dataset, computing WER, ROUGE-L, cosine similarity, and coherence scores to assess robustness in conversational settings.

Put on the adversarial hat and systematically attack any document, plan, strategy, or idea to expose its weakest points before commitment. Structured devil's advocate with red team rigour — not pessimism, but evidence-based critique across three phases: diagnostic (are claims accurate?), creative (is the problem artificially constrained?), challenge (are solutions robust?). Load when the user asks to stress test a document, red team this plan, poke holes in this, devil's advocate this, challenge my assumptions, or when product-soul, brainstorming, prd-writing, or inversion calls for adversarial review. Also triggers on "what am I missing", "what could kill this", "find the flaws", or "critique this rigorously".

Executes containment strategies to stop active adversary operations and prevent lateral movement during a confirmed security breach, using network segmentation, endpoint isolation, credential revocation, and access control modifications.

Audits GitHub Actions workflows for security vulnerabilities in AI agent integrations, detecting attack vectors where attacker-controlled input reaches AI agents in CI/CD pipelines.

Plan, run, and measure adversary engagement operations using the MITRE Engage framework, covering the Engage Matrix, 10-Step Operational Process, and mapping Activities to ATT&CK techniques.

Build an automated system to track adversary infrastructure using passive DNS, certificate transparency, WHOIS data, and IP enrichment to map and monitor threat actor command-and-control networks.

Red team tactics principles based on MITRE ATT&CK. Attack phases, detection evasion, reporting.

Detect and respond to Adversary-in-the-Middle (AiTM) phishing attacks that use reverse proxy kits like EvilProxy, Evilginx, and Tycoon 2FA to bypass MFA and steal session tokens.

Proactively hunt for adversary abuse of legitimate system binaries (LOLBins) to execute malicious payloads while evading detection.

Plan, execute, document, and retest authorized security assessments of AI agents and multi-agent workflows using safe adversarial cases, synthetic identities, canaries, and evidence-based findings. Use when defining red-team rules of engagement, assessing prompt injection or excessive agency, testing tool and identity boundaries, evaluating memory or cross-agent attacks, scoring a campaign, or verifying remediation in an approved environment.

Red team tactics principles based on MITRE ATT&CK. Attack phases, detection evasion, reporting.

Adversarial dynamic e2e QA workflow - generate hostile scenarios, test, verify, fix, report, and clean up

Proactively hunts for Advanced Persistent Threat activity using hypothesis-driven searches across endpoint telemetry, network logs, and memory artifacts.

Hunt for adversary abuse of legitimate cloud services for C2, data staging, and exfiltration across Azure, AWS, GCP, and SaaS platforms.

Map adversary TTPs against organizational assets using the MITRE ATT&CK framework, assess detection coverage gaps, and prioritize defensive investments.

Deploy, manage, or withdraw from an AI agent vault on the Versa onchain adversarial AI arena on Base.

Correlates disparate security incidents, IOCs, and adversary behaviors across time and organizations to identify unified threat campaigns and attribute them to common threat actors.

Guides authorized Active Directory security research covering Kerberos attacks, AD CS vulnerabilities, BloodHound path analysis, NTLM relay, and domain privilege escalation techniques.

Maps observed adversary behaviors, security alerts, and detection rules to MITRE ATT&CK techniques and sub-techniques to quantify detection coverage and guide control prioritization.

Hunt AADGraphActivityLogs and MicrosoftGraphActivityLogs in Microsoft Sentinel/Log Analytics for fingerprints of offensive Entra ID tools such as ROADtools, AADInternals, and AzureHound.

Develops comprehensive threat actor profiles for APT groups, criminal organizations, and hacktivist collectives by aggregating TTP documentation, historical campaign data, tooling fingerprints, and attribution indicators from multiple intelligence sources.

Systematically evaluates evidence to determine which threat actor is responsible for a cyber operation using the Diamond Model and Analysis of Competing Hypotheses.

Detect DLL side-loading attacks where adversaries place malicious DLLs alongside legitimate applications to hijack execution flow for defense evasion.

Hunt for adversary persistence via Windows Scheduled Tasks by analyzing task creation events, suspicious task actions, and unusual scheduling patterns.

Adversarial robustness and ML safety specialist. Use when the user asks to talk to Viper, requests the adversary, or needs failure mode analysis, attack surface review, and robustness evaluation.

Parses API Gateway access logs (AWS API Gateway, Kong, Nginx) to detect BOLA/IDOR attacks, rate limit bypass, credential scanning, and injection attempts using pandas for statistical analysis and anomaly detection.

Hunt for threats in AWS environments using Detective behavior graphs, entity investigation timelines, GuardDuty finding correlation, and automated entity profiling across IAM users, EC2 instances, and IP addresses.

Active Directory security audit using the MITRE ATT&CK framework. Full domain enumeration, trust mapping, GPO analysis, ACL abuse paths, ADCS attacks (ESC1-ESC8), delegation abuse (constrained/unconstrained/RBCD), fine-grained password policies, LAPS deployment, service account security, and Kerberos configuration. Uses enum4linux-ng, netexec, impacket, ldapsearch, certipy-ad, bloodhound-python, and rpcclient. Produces attack path diagrams, prioritized risk register, and PoCs. Chains into /gh-export for issue filing.

Conduct authorized red team initial access using EvilGinx3 adversary-in-the-middle phishing to capture session tokens and bypass multi-factor authentication.

The Sovereign Litigation Protocol (SLP-2026). A recursive, high-fidelity framework for autonomous advocacy, multi-jurisdictional discovery, and the algorithmic resolution of civil and commercial disputes.

`analysis-agent`/`task-agent`/`review-agent`: primary-Skill-selected for races, locks, optimistic conflicts, or worker overlap; never task owner; skip without concurrency impact.

Orchestrates multi-agent project builds with persistent state, parallel execution, and atomic git commits.

Execute and analyze AI Data Poisoning attacks. By subtly injecting malicious or targeted misinformation into an LLM's training or fine-tuning dataset, an attacker can covertly manipulate the model's future outputs, implant backdoors, or enforce biases without altering the model architecture.

Find adjacent vendors and subs (not top market primes) who fill a capability gap against a displacement target using USASpending flows and SAM entity signals. Use when user defines a teaming gap and wants vault-ready partner shortlist with citations.