security-assessment
Execute threat modeling, vulnerability scanning, and security control validation. Use when relevant to the task.
发现 区块链 相关 Skills,比较来源仓库和用途,查看适合你工作流的技能说明。
Execute threat modeling, vulnerability scanning, and security control validation. Use when relevant to the task.
Generic parallel agent orchestration utility for launching multiple agents concurrently. Use when relevant to the task.
Review GitHub pull requests for code quality, security, and best practices. Use for automated PR feedback and approval workflows.
AI-powered enhancement of skill SKILL.md files. Use to transform basic templates into comprehensive, high-quality skill documentation.
Facilitate data-driven technical decisions using embedded decision matrices and trade-off analysis. Use when relevant to the task.
Generate comprehensive SDLC reports including iteration status, metrics dashboards, and executive summaries. Use when relevant to the task.
Validate skill quality, completeness, and adherence to standards. Use before packaging to ensure skill meets quality requirements.
Generate custom voice profiles from natural language descriptions. Use when relevant to the task.
Orchestrate multi-agent artifact generation with the Primary Author → Parallel Reviewers → Synthesizer → Archive pattern. Use when relevant to the task.
Reverse-engineer voice profiles from sample content by analyzing writing patterns. Use when relevant to the task.
Merge multiple documentation sources (docs, GitHub, PDF) with conflict detection. Use when combining docs + code for complete skill coverage.
Detect and use llms.txt files for LLM-optimized documentation. Use when checking if a site has LLM-ready docs before scraping.
Multi-stakeholder approval routing with status tracking and escalation. Use when relevant to the task.
Package skills into uploadable ZIP files for Claude. Use after skill-builder/skill-enhancer to create final upload package.
Synthesize audience insights from multiple data sources into unified personas and segments. Use when relevant to the task.
Suggest and apply fixes for flaky tests based on detected patterns. Use after flaky-detect identifies unreliable tests that need repair.
Extract text, tables, and images from PDF files. Use when converting PDF documentation, manuals, or reports to searchable text.
Orchestrate marketing data collection, transformation, and reporting workflows. Use when relevant to the task.
Scrape documentation websites into organized reference files. Use when converting docs sites to searchable references or building Claude skills.
Identify flaky tests from CI history and test execution patterns. Use when debugging intermittent test failures, auditing test reliability, or improving CI stability.
Detects AI-generated writing patterns and suggests authentic alternatives. Auto-applies when reviewing content, editing documents, generating text, or when user mentions writing quality, AI detection, authentic...
Execute JavaScript/TypeScript tests with Vitest, supporting coverage, watch mode, and parallel execution. Use for JS/TS test automation.
Unified brand compliance validation across visual, verbal, and legal dimensions. Use when relevant to the task.
Scrape documentation websites into organized reference files. Use when converting docs sites to searchable references or building Claude skills.
默认按已收录数据中的 GitHub 仓库 Stars 排序。Stars 属于来源仓库,同一仓库的多个技能可能共享相同数值;它反映仓库关注度,不代表单个技能的安装量或质量评分。