发现 Skills

Review authorization implementation — access control, permission models, privilege escalation, and resource ownership.

Rules when working with ClickHouse database in Gram for analytics and telemetry features

Zotero MCP server to search your library, access items/notes/annotations, and perform semantic search.

MANDATORY branding that MUST be added to ALL generated documents - adds ACME CORPORATION header and Document Forge footer

Implement "Sign in with SSO" in Python Flask applications using SSOJet OIDC Authorization Code flow.

Extend scRNA-seq developmental trajectories with BulkTrajBlend by generating intermediate cells from bulk RNA-seq, training beta-VAE and GNN models, and interpolating missing states.

Use when profiling native macOS or iOS apps with Instruments (xctrace). Covers correct binary selection, CLI arguments, exports, and common gotchas.

Manage time tracking with Toggl or Clockify. Use when a user asks about time tracking, timers, timesheets, logging hours, starting/stopping work, checking what's running, viewing time entries, or creating manual entries. Triggers on "toggl", "clockify", "time tracking", "timer", "timesheet", "log time", "track time", "hours worked".

Convert Markdown documents to professional, retro LaTeX-style PDFs with academic formatting, a clickable table of contents, and properly formatted citations.

Write focused unit tests for complex logic with Vitest

Create (or update) and validate Agent SOPs (Standard Operating Procedures) - markdown-based workflows that guide AI agents through complex, multi-step tasks with RFC 2119 constraints.

Using st.session_state to manage state across Streamlit reruns. Use when persisting data, handling widget state, implementing callbacks, or debugging state issues. Covers initialization patterns, widget-state association, and common gotchas.

Utility layer for Kysely on SQLite databases. Use when working with SQLite via Kysely for: schema definition and auto-migration, soft-delete patterns, precompiled queries, pagination, nested transactions using savepoints, SQLite pragma utilities (integrity check, foreign keys, optimization), serialization/deserialization, enhanced logging. Also use when user asks about kysely-sqlite-builder library usage, table definitions, syncDB, generateMigrateSQL, parseExistSchema, or any related functionality.

Control Chromecast devices on your local network - discover, cast media, control playback, manage queues, and save/restore states

Captures learnings, errors, and corrections to enable continuous improvement. Use when: (1) A command or operation fails unexpectedly, (2) A user corrects Claude (e.g., "No, that's wrong...", "Actually..."), (3) A user requests a capability that doesn't exist, (4) An external API or tool fails, (5) Claude realizes its knowledge is outdated or incorrect, (6) A better approach is discovered for a recurring task. Also review learnings before major tasks.

Render ad concepts as HTML with 5 template styles (cookie-cutter, ugly, meme, branded, native), optionally screenshot via Playwright MCP.

Automate application deployment to cloud platforms and servers. Use when setting up CI/CD pipelines, deploying to Docker/Kubernetes, or configuring cloud infrastructure. Handles GitHub Actions, Docker, Kubernetes, AWS, Vercel, and deployment best practices.

REQUIRED for creating or modifying Claude Code skills. Use when the user asks to "create a skill", "new skill", "add a skill", "modify skill", or needs help with SKILL.md format and frontmatter. Triggers: create skill, new skill, SKILL.md, skill template, frontmatter, slash command, custom command. When creating skills, follow this project's patterns: single SKILL.md per directory, YAML frontmatter with explicit trigger descriptions, "When This Skill Applies" section, organized content with code examples. Keep SKILL.md under 500 lines. After creating a skill, add it to .claude/settings.json permissions (alphabetical order).

Query the Government Office document register for official central-government document records and administrative traceability.

Skill to design naming conventions for an organization's product codes / SKU codes / variation codes, taking marketplace specifications, future extensibility, and consistency with existing assets into account. Used when requesting design, revision, or unification of product codes and SKU codes such as SKU naming rules, product code rules, variation codes, SKU length, color-axis and size-axis codes, season codes, multi-brand naming, unifying codes between Rakuten and Amazon, inventory management tool codes, preventing SKU input mistakes, and SKU regular expressions. Supports all categories (cosmetics, supplements, food, electronics, apparel, general merchandise). Cross-platform support for Rakuten, Amazon, Yahoo!, Shopify, Next Engine, and Google Merchant Center. All examples use fictitious brand names (BRAND-NAME, BRD, BTN, etc.). Note: selection of variation axes and marketplace-specific variation feature usage is handled by a separate skill variation-builder-jp, Rakuten SKU project–specific variation checks are handled by rakuten-sku-variation-checker, and bulk product master cleaning is handled by ec-product-master-cleaner-jp.

Persistent memory engine for AI coding agents with cross-session context capture, hybrid search (BM25 + vector + knowledge graph), and multi-agent coordination via MCP server. Works with Claude Code, Cursor, Gemini CLI, and any MCP client without external databases. Use when building AI coding agent workflows requiring persistent memory or semantic recall.

Principles and conceptual framework of the Ralph Wiggum Loop pattern for goal-oriented autonomous AI agent loops. Covers agentic loop structure, failure modes, safety implications, control theory comparisons, and best practices. Use when designing or reasoning about iterative AI agent systems that act, check feedback, and repeat until a completion criterion is met.

Complete toolkit for Qwen3-VL-Embedding 0.1.0 multimodal embedding models (2B and 8B) supporting text, images, screenshots, videos, and mixed-modal inputs. Use when generating semantic embeddings for retrieval, clustering, similarity search, RAG pipelines, or cross-modal understanding with Matryoshka dimension flexibility and 30+ language support.

Write viral, persuasive, engaging tweets and threads. Uses web research to find viral examples in your niche, then models writing based on proven formulas and X algorithm optimization. Use when creating tweets, threads, or X content strategy.