Skill UI

A comprehensive methodology for discovering potential software vulnerabilities. It outlines advanced techniques including static analysis, dynamic analysis, fuzzing, taint tracking, and modern attack surface mapping (e.g., kernel, drivers). Ideal for professional code audits, bug bounty hunting, and building robust vulnerability research pipelines.

This guide provides comprehensive best practices for configuring Palantir Foundry integrations across development, staging, and production environments. It details how to manage environment-specific credentials, hostnames, and scope policies using structured Python code and platform-specific deployment manifests (Docker Compose, Kubernetes, gcloud). Use this skill when implementing resilient, multi-tier SaaS deployments that require strict credential and configuration separation.

This advanced playbook details expert techniques for format string exploitation. It covers arbitrary stack reading (%p), memory writing (%n), and exploiting weaknesses like GOT/hook overwrites. Essential for CTF challenges and binary vulnerability analysis to achieve reliable Remote Code Execution (RCE).

A comprehensive playbook detailing expert techniques to escape multiple types of restricted environments, including Python pyjail, Docker containers, seccomp filters, and browser rendering processes. It covers advanced exploitation vectors like namespace escapes, kernel exploits, and built-in function bypasses for achieving code execution.

This expert playbook details advanced techniques for exploiting stack-based vulnerabilities, including classic buffer overflows, Return-Oriented Programming (ROP), ret2libc, and specialized gadgets like ret2csu and SROP. It provides comprehensive guidance on bypassing modern memory mitigations (NX, ASLR, PIE) to hijack program control flow and achieve reliable remote code execution in userland binaries.

This skill facilitates a rigorous, documentation-anchored design review session. It challenges a proposed plan by cross-referencing it against the project's established domain glossary (CONTEXT.md), historical architectural decisions (ADRs), and the current codebase. It forces the user to resolve ambiguities, define precise terminology, and validate design trade-offs using concrete scenarios, ensuring all decisions are captured and documented inline.

A structured workflow designed to prevent 'reinventing the wheel.' Before writing custom code, this skill guides the user through analyzing needs, conducting parallel searches across multiple channels (npm, PyPI, GitHub, MCP), evaluating candidate solutions based on maintainability and community support, and making a decision (Adopt, Extend, or Custom Build). It ensures that solutions leverage existing, proven libraries and patterns.

Provides comprehensive guidelines for authors submitting papers to Acta Automatica Sinica (AAS). It details how to organize and present anonymous, reproducible code, data, and experimental artifacts to enhance the credibility and rigor of research in control, automation, and pattern recognition, supporting the double-blind review process.

This comprehensive guide outlines best practices for achieving rigorous scientific reproducibility, especially for research involving storage systems and hardware dependencies. It mandates documenting not only the code but also the specific device models, firmware versions, operating states, and full measurement protocols to ensure verifiable results by peer readers.

This guide outlines the rigorous standards for achieving robust reproducibility in robotics research papers, especially those involving physical hardware. It details a two-tier approach, defining results as either fully *rerunnable* (code/simulation) or fully *auditable* (hardware logs/protocols). Key requirements include creating comprehensive hardware specification ledgers, logging detailed sensor data (rosbags), separating deterministic from stochastic claims, and structuring the entire repository for peer review.

A comprehensive guide for structuring the evidence bundle required for robotics research submissions. It details the critical components—including hardware ledgers, time-stamped logs, configuration files, code, and data—necessary to ensure that scientific claims are fully auditable and reproducible, addressing the needs of skeptical reviewers.

A comprehensive guide for authors preparing artifacts (code, data, models, etc.) for submission to the Journal of Computer-Aided Design & Computer Graphics (JCAD&CG). It details best practices for ensuring research reproducibility, writing honest availability statements, and managing artifacts during the double-blind review process, focusing on ethical standards regardless of specific journal policy.