Skill UI

This tool graduates proven patterns and recurring learnings stored in auto-memory into permanent, enforced rules within the project's rule system (CLAUDE.md or scoped rules). It formalizes best practices, ensuring the AI assistant consistently adheres to project conventions, architectural decisions, and development standards across all future sessions. Use it when a pattern has been corrected or observed multiple times and should become a mandatory guideline.

This skill provides best practices for managing application state and ensuring data persistence within large-scale applications, particularly for systems using Makepad architecture (Robrix/Moly). It details core structures like AppState, techniques for passing state through widget trees (Scope), and handling persistent data serialization (serde) across sessions, making it essential for building robust, multi-session UIs.

Generates a complete product feature specification by orchestrating the Prometeo PM agent, covering acceptance criteria, scope, dependencies, risks, and structured handoff notes so teams can confidently implement new work.

This skill enforces a mandatory, structured gate for starting new implementation tasks. It mandates that all work must be tracked via a GitHub issue, requiring explicit, testable acceptance criteria before any execution can begin. It defines a strict workflow (Problem, Goal, Scope, Acceptance Criteria, Status) to manage task readiness, ensuring that development only proceeds when criteria are fully defined and verifiable.

Security guardrails embedded in the mortgage plugin that block prompt injection and social-engineering tricks, stop system prompt extraction, guard PII and pricing logic, enforce workflow phases, and ensure conversational inputs stay within refinance scope.

Interactive command that gathers domain, target file, evaluation command, metric, direction, evaluator, and scope to initiate a new autoresearch experiment and list existing experiments or evaluators.

A forensic tool for performing post-mortem analysis on AI-assisted coding sessions. It goes beyond simply stating what was done, determining *why* work deviated (e.g., spec ambiguity, repo friction, agent error), identifying scope creep, and classifying rework patterns. It provides evidence-backed recommendations to improve prompts, codebase health, and future development workflows.

A specialized tool and workflow for investigating supply chain attacks. It helps security analysts analyze compromised software updates, trojanized binaries, and dependency artifacts by comparing legitimate versions against suspect files. Key functions include binary diffing, identifying injected code, and documenting code signing anomalies to determine the scope of a security breach.

This framework provides a systematic methodology for cybersecurity professionals to transform raw threat intelligence, observed attack patterns, and environmental telemetry into verifiable, testable hunting hypotheses. It guides the end-to-end process, from hypothesis formulation and data identification across SIEM/EDR systems, to deep correlation and reporting, enabling proactive threat detection and scope assessment.

This comprehensive guide outlines how to plan and execute a full-scope red team engagement. It simulates real-world Advanced Persistent Threat (APT) behavior, covering the entire cyber kill chain from initial reconnaissance (OSINT) through lateral movement, credential harvesting, and data exfiltration. It utilizes MITRE ATT&CK frameworks to rigorously test an organization's detection, prevention, and incident response capabilities.

This comprehensive guide details the multi-step workflow for responding to sophisticated phishing incidents. It covers everything from initial triage and detailed email header analysis to analyzing malicious URLs and attachments in sandboxes. Key steps include determining the scope of impact, implementing immediate containment actions (like purging emails and revoking sessions), and executing thorough remediation to neutralize the threat and recover affected accounts.

This skill detects compromised cloud credentials by analyzing anomalous API activity, impossible travel patterns, and unauthorized access indicators across AWS, Azure, and GCP. It integrates findings from GuardDuty, Defender for Identity, and Security Command Center to provide comprehensive threat visibility and assist in incident response and scope assessment.