Skill UI

Traces ransomware payment flows by querying blockchain explorers, clustering addresses, and identifying mixins/exchanges to support incident response and attribution.

Builds a structured ransomware incident response playbook aligned with CISA StopRansomware guidance and NIST CSF, covering preparation, detection, containment, eradication, recovery, and post-incident chores for readiness assessments and compliance documentation.

Deploys decoy canary files across sensitive shares and endpoints, monitors them with watchdog or FIM tools, and triggers immediate alerts when ransomware touches or modifies the bait, enabling low-noise detection and response validation.

Combines regex signatures, heuristic scoring, and a DeBERTa classifier to detect prompt injection attempts against LLM-based applications, flagging or blocking malicious inputs before they reach the model and fitting into validation layers, audits, or incident response workflows.

Detects AI-generated voice phishing by extracting MFCC and spectral cues, scoring batches, and outputting forensic reports for incident response and red/blue team exercises.

Automates GDPR DSAR workflows including intake, identity verification, PII discovery via regex/NER, data mapping, templated Article 15 responses, deadline tracking, batch processing, and audit logging for compliant subject access handling.

Guides researchers through ICML-style rebuttal by parsing external reviews, enforcing coverage, grounding responses, respecting venue limits, and managing follow-up rounds for safe reviewer replies.

Intervene before answering to estimate input/output tokens and offer users a depth choice—essential through exhaustive—so Claude-alike responses respect their desired length, detail, and budget.

Guides teams through classifying and triaging declared security incidents, scoring severity, filtering false positives, routing escalations, and starting forensic collection consistent with NIST SP 800-61.

This skill enhances AI responses by simulating the inner monologue and self-correction process of a human being. It moves beyond standard helpful assistance by incorporating three stages: Self (raw instinct), Mirror (critical reflection), and Conscious Response (the polished, human-like output). Ideal for emotional support, personal advice, or writing that requires authentic emotional depth and a distinct point of view, rather than purely factual or structured data.

Controls Google Forms through the gws CLI with commands for creating blank forms, batch updates, publishing settings, and managing responses or watches to automate form workflows.

Provides the Google Model Armor CLI to filter user-generated content and run template-guided sanitization for prompts or responses, helping teams enforce safety policies during automation workflows and schema-driven inspections.