Skill UI

An advanced reliability wrapper for AI agents that monitors, manages, and recovers from tool call failures. It automatically handles common issues such as truncated JSON responses, API timeouts, rate limits, and mid-chain process breakdowns. By implementing sophisticated failure classification, checkpointing, and learning mechanisms, Guardian ensures complex, multi-step agent workflows remain stable and resilient.

A comprehensive skill for performing deep security audits on Ethereum smart contracts. Utilizes both static analysis (Slither) and symbolic execution (Mythril) to detect critical vulnerabilities such as reentrancy, integer overflows, and access control issues. This process is crucial for ensuring the security and immutability of financial assets before mainnet deployment.

This guide details the comprehensive process of using the Linux Audit framework (auditd) to detect advanced security threats. It covers deploying specific rules to monitor critical system files, detecting unauthorized privilege escalation, process injection, and suspicious system calls. Users can then leverage ausearch and aureport to efficiently query, reconstruct timelines, and summarize findings for forensic analysis and compliance auditing.

This guide details the process of performing static analysis on potentially malicious PDF documents. It utilizes advanced tools like peepdf, pdfid, and pdf-parser to extract embedded JavaScript, shellcode, and suspicious objects. Use this method when triaging suspicious attachments from phishing emails or conducting deep forensic examination of weaponized document artifacts to generate actionable Indicators of Compromise (IOCs).

This skill automates the execution of suspicious malware samples within a secure, isolated Cuckoo Sandbox environment. It observes comprehensive runtime behaviors, including process creation chains, file system tampering, registry modifications, network communications, and API calls. It is essential for generating detailed behavioral reports, extracting Indicators of Compromise (IOCs), and classifying unknown malware samples.

This skill provides a comprehensive workflow for detecting and analyzing malware persistence mechanisms on Windows systems. It automates the use of Sysinternals Autoruns via scripting to scan hundreds of Auto-Start Extensibility Points (ASEPs), including registry keys, services, scheduled tasks, and drivers. The process flags suspicious entries like unsigned binaries, suspicious paths, or dangerous command-line arguments, making it an essential tool for security incident response and threat hunting.

Walks through analyzing RAM dumps with Volatility 3/2 to uncover injected code, hidden processes, network activity, and credentials when responding to compromised hosts.

This skill outlines the comprehensive process of memory forensics on compromised Linux systems. It guides users through acquiring volatile memory using the LiME kernel module and analyzing the resulting image with the Volatility 3 framework. Key artifacts extracted include process lists, network connections, bash history, and loaded kernel modules, making it an essential technique for incident response, threat hunting, and security investigations.

This skill provides a comprehensive workflow for identifying, unpacking, and analyzing packed malware samples, especially those utilizing the UPX format. It details methods for standard UPX decompression and advanced techniques for repairing modified PE headers or magic bytes that prevent automated extraction. The process supports preparing samples for deep static analysis and dynamic debugging.

This skill automates the scholarly process of refining a generated academic paper. It executes an iterative loop (Review → Fix → Recompile) using advanced LLMs (like GPT-5.5) over multiple rounds. It is designed not for primary research, but specifically for polishing writing quality, correcting theoretical inconsistencies, improving presentation, and enhancing narrative flow after the initial draft is complete. It helps elevate a paper from a draft to a polished submission.

This skill automates the research improvement process by creating a closed loop: Review -> Implement Fixes -> Re-review. It leverages any OpenAI-compatible LLM API to iteratively refine research work until a positive assessment is reached or a maximum round limit is hit. Ideal for academic or technical project refinement.

This advanced skill automates the rigorous, iterative review and refinement of complex research topics or papers. It runs multi-round cycles, using an external LLM (MiniMax API) to provide continuous, constructive feedback. The process guides the user through review, fixes, and re-review stages until a positive assessment is achieved or the maximum round limit is reached. It includes robust state persistence for long-running academic workflows.