发现 Skills

Use when you want to capture project-specific patterns as reusable skills - generates custom skills based on codebase patterns, common workflows, and team conventions discovered during analysis. Creates skills for patterns, testing, deployment, debugging, and setup. Do NOT use for generic skills that would work across any project - those belong in popkit core, not project-specific generation.

Deep reading collaborative system using multi-layered AI agents to help transform articles from 'read' to 'understood' to 'mastered', with actionable plans. Use when users need to deeply understand complex articles/papers, organize reading notes systematically, think critically about content, discover hidden logical issues and assumptions, or transform knowledge into action plans. Trigger keywords: deep reading, critical thinking, reading notes, article analysis, Socratic questioning, action plan

Comprehensive PDF manipulation toolkit for extracting text and tables, creating new PDFs, merging/splitting documents, and handling forms. Use when Claude needs to fill a PDF form or programmatically process, generate, or analyze PDF documents at scale.

Master binary analysis patterns including disassembly, decompilation, control flow analysis, and code pattern recognition. Use when analyzing executables, understanding compiled code, or performing static analysis on binaries.

A software security skill that integrates with Project CodeGuard to help AI coding agents write secure code and prevent common vulnerabilities. Use this skill when writing, reviewing, or modifying code to ensure secure-by-default practices are followed.

Derive security requirements from threat models and business context. Use when translating threats into actionable requirements, creating security user stories, or building security test cases.

Git commit guidelines with conventional format, safety checks, and granular strategy. Use when making commits, staging changes, or writing commit messages.

Find, install, and configure OpenCode plugins from the catalog or community. Use proactively when a user asks about plugins, requests new capabilities, or mentions extending OpenCode functionality. Examples: - user: "Is there a plugin for Tailwind CSS?" → list catalog, read Tailwind plugin details, install if available - user: "How do I add a custom slash command?" → suggest a command-creator skill or guide through opencode.json setup - user: "What plugins are available for database work?" → list catalog, filter for database-related plugins - user: "Install the playwright plugin" → read plugin file, add to opencode.json, verify installation

Guide competency framework development and operation. Use when building training that produces capability, when existing training doesn't produce competence, when structuring knowledge for multiple audiences, or when setting up feedback loops to surface gaps.

Diagnose weak endings, rushed resolutions, and arbitrary conclusions. Use when stories build well but end disappointingly, when climax feels unearned, or when resolution doesn't complete character arcs.

Generate speculative fiction stories about systemic exploitation by collapsing comfortable moral distances. Use when exploring how privilege and harm are connected, when writing about systems that export consequences, or when you want stories where innocence becomes impossible.

Design multi-generational societal evolution for science fiction settings. Use when creating civilizations that diverge from baseline humanity, when exploring how environments shape cultures over generations, or when worldbuilding requires deep time development.

Generate stories where ordinary people become crucial through their structural position in systems. Use when you want protagonists who aren't chosen ones but accidental pivots, when mundane jobs should reveal conspiracies, or when you need structurally inevitable involvement rather than coincidence.

Diagnose requirements problems and guide discovery of real needs and constraints

Transform clichéd story elements by pushing along the emotional vector toward statistical edges. Use when first instincts are too predictable, when elements feel generic, or when you need the core methodology for avoiding statistical-center defaults.

Generate publication-quality academic figures, framework diagrams, and pipeline illustrations from paper method text and figure captions. Use when the user asks to draw a figure for a paper, create a method overview diagram, or visualize a research pipeline/framework/architecture.

Convert Metabase questions, models, and metrics into KTX Semantic Layer source definitions. Covers result-metadata to KSL column type mapping, FK/PK detection, near-duplicate deduplication, pre-aggregation decomposition, join-graph connectivity, and how to react to priorProvenance from earlier ingest syncs. Load when the WorkUnit contains `cards/<id>.json` files under a Metabase bundle.

Read and extract full content from WeChat Official Account article URLs (mp.weixin.qq.com), including title/author/body, and return clean Markdown text. Use when user asks to parse, summarize, archive, or convert a WeChat article link into readable text.

API testing skill using REST Assured for Java, covering request specifications, response validation, authentication, JSON schema validation, and serialization.

Use this skill whenever the user wants to explore, profile, or understand a tabular dataset for general data science or ML purposes. This includes requests to perform exploratory data analysis (EDA), profile a dataset, understand data quality, find patterns or anomalies, check distributions, assess feature relationships, or prepare data understanding before modeling. Trigger when the user uploads a CSV, Excel, Parquet, or other tabular data file and asks anything like “analyze this data”, “what does this dataset look like”, “EDA”, “data profiling”, “explore this”, “tell me about this data”, “check data quality”, or “summarize this dataset”. Also trigger when the user explicitly mentions EDA or exploratory analysis even without uploading a file yet. Do NOT trigger for pure modeling/ML tasks, dashboard building, ETL pipeline creation, or any demand/sales forecasting, time series forecasting, or retail/supply chain analysis — use the forecasting skill for those instead.

Applies Anthropic's official brand colors and typography to any sort of artifact that may benefit from having Anthropic's look-and-feel. Use it when brand colors or style guidelines, visual formatting, or company design standards apply.

Apple documentation research, Xcode MCP (mcpbridge) setup/tools/reference, App Store Connect MCP, WWDC transcript extraction, sosumi.ai docs, bundled compiler diagnostics

EXPERIMENTAL: Three-layer parallel meta-cognition analysis. Triggers on: /meta-parallel, parallel analysis

Use when a task requires a spec-driven workflow: draft/refresh PRD + TECH_SPEC + ACTION_PLAN + tasks, link TECH_SPEC in tasks/index.json, and run docs-review before implementation.