Skill UI

Autonomously improve a generated paper through multiple review-fix-recompile rounds with fresh LLM reviewer feedback, fixing inconsistencies and polishing presentation.

Autonomously iterates review, fix, and re-review cycles until an independent reviewer approves the work or the maximum round limit is reached. Supports multiple reviewer backends and difficulty levels for research and code improvement.

This skill enables an autonomous research review loop using any OpenAI-compatible LLM API. It iterates through review, implementation, and re-review phases until a positive assessment is achieved or maximum rounds are reached. Configure via MCP server or environment variables. Trigger with specific commands to improve research quality automatically.

This skill implements an autonomous, multi-round research review loop, simulating the rigorous peer-review process of top-tier conferences. It iteratively reviews, identifies weaknesses, suggests fixes, and re-reviews a research document using the MiniMax API. The system manages state persistence across multiple rounds, ensuring continuous improvement until a 'ready' assessment is achieved or the maximum round limit is reached. Ideal for academic writing, thesis refinement, or comprehensive project documentation.

Establishes a repeatable, structured operational process for managing Microsoft Patch Tuesday security updates. The workflow guides users through advisory review, vulnerability triaging (including zero-days), phased testing, and deployment across major platforms (Windows, Azure, Office). Ideal for maturing patch management programs and defining risk-based remediation SLAs.

A structured guide for conducting blameless post-mortem reviews after any security incident. This process helps identify root causes using techniques like the 5 Whys, calculates critical metrics (MTTD, MTTR), and generates actionable recommendations to continuously improve incident response playbooks and procedures.

A comprehensive guide and framework for detecting attacker lateral movement within enterprise networks. It details the use of network flow analysis (Zeek), authentication log review (Windows Event Logs, Kerberos), and dedicated SIEM correlation rules (Splunk, Elastic) to identify techniques like Pass-the-Hash and RDP hopping. Ideal for threat hunters and security engineers building detection pipelines.

This skill provides a comprehensive methodology for discovering and exploiting broken link hijacking vulnerabilities. By spidering a target site and extracting all external scripts, domains, and references, the skill checks the DNS and domain registration status. It is crucial for conducting subdomain takeover testing, supply-chain security reviews of third-party resources, and advanced bug bounty hunting.

A robust internal utility for integrating system status updates with Feishu/Lark. It allows other skills to send structured notifications (e.g., experiment complete, checkpoint waiting, review scores) via webhooks. Supports both 'Push only' (fire-and-forget) and 'Interactive' (bidirectional polling) modes, guaranteeing zero side effects if configuration is missing.

This skill orchestrates a complete automated pipeline for discovering research ideas. It chains literature review, brainstorming, novelty checks, and pilot testing with GPU constraints to produce validated concepts and experiment plans.

This skill automates the discovery of robotics research ideas by orchestrating a literature survey, idea generation, novelty check, and critical review workflow. It focuses on embodied AI, sim2real, and benchmark-grounded concepts for manipulation, locomotion, and navigation. The pipeline ensures ideas are feasible, falsifiable, and aligned with robotics constraints like simulation-first approaches and hardware availability.

This guide details the implementation of Microsoft Entra Privileged Identity Management (PIM), enforcing Just-in-Time (JIT) role activation, mandatory approval workflows, and periodic access reviews for critical Azure AD and resource roles. It aligns with Zero Trust principles by eliminating standing privileges.