发现 Skills

Integrate Gemini API with correct current SDK (@google/genai v1.27+, NOT deprecated @google/generative-ai). Supports text generation, multimodal (images/video/audio/PDFs), function calling, and thinking mode. 1M input tokens. Use when: integrating Gemini API, implementing multimodal AI, using thinking mode for reasoning, function calling with parallel execution, streaming responses, deploying to Cloudflare Workers, building chat, or troubleshooting SDK deprecation, context window, model not found, function calling, or multimodal format errors. Keywords: gemini api, @google/genai, gemini-2.5-pro, gemini-2.5-flash, gemini-2.5-flash-lite, gemini-3-pro-preview, multimodal gemini, thinking mode, google ai, genai sdk, function calling gemini, streaming gemini, gemini vision, gemini video, gemini audio, gemini pdf, system instructions, multi-turn chat, DEPRECATED @google/generative-ai, gemini context window, gemini models 2025, gemini 1m tokens, gemini tool use, parallel function calling, compositional function calling, gemini 3

Git conventions for branches, commits, and pull requests. Load this Skill whenever creating branches, writing commit messages, or opening PRs. Trigger on any git workflow operation — commits, branches, pull requests, or merge operations.

Third sentry checkpoint of the Reviewer corps. Focused on finding vulnerabilities — if a plaintext password is discovered, the submission is immediately rejected.

Security awareness training is the human layer of phishing defense. An effective anti-phishing training program combines regular simulations, interactive learning modules, metric tracking, and positive reinforcement.

Performs comprehensive security assessments of IoT devices and their ecosystems by testing hardware interfaces, firmware, network communications, cloud APIs, and companion mobile applications. The tester uses firmware extraction and analysis, hardware debugging via UART and JTAG, network protocol analysis, and runtime exploitation to identify vulnerabilities across all layers of the IoT stack. Activates for requests involving IoT security testing, embedded device assessment, firmware security analysis, or smart device penetration testing.

name: rag-knowledge-base description: Build RAG (Retrieval-Augmented Generation) knowledge bases for businesses — turn documents, SOPs, policies, and product manuals into AI assistants that answer questions with source citations and confidence scores. The pipeline ingests and preprocesses corpora, generates embeddings, stores vectors in scalable vector databases, and orchestrates retrieval + LLM components (e.g., LlamaIndex, OpenAI) to produce citation-backed responses. Key capabilities include document connectors, chunking with metadata preservation, vector search, citation-aware answer generation, monitoring, periodic reindexing, and integration with chat, web, and Slack. Designed to reduce support load, deliver more grounded answers (fewer hallucinations), speed knowledge access, and provide a clear commercial path with build and maintenance SLAs.

Guides the migration of TypeScript projects from AlgoKit Utils v9 to v10. Use when upgrading dependencies, refactoring legacy function-based code to the new fluent AlgorandClient API, or resolving deprecation warnings related to v10 architectural changes. Triggers include "upgrade algokit-utils", "migrate from v9 to v10", "use AlgorandClient", "replace getAlgoClient", "refactor to fluent API".

Planning work in small, known-good increments. Use when starting significant work or breaking down complex tasks.

Full-text search across employees, chats, and messages. Use when you need to: find a message by text, find a chat by name, find an employee by name. DO NOT use for: viewing the list of employees (→ pachca-users), viewing the list of chats (→ pachca-chats).

Complete automated onboarding: analyzes codebase, creates package, and generates standards & commands via CLI. Automatic package creation when none exist, user selection when packages are available.

Use when working with FS25 textures, DDS formats, normal maps, specular maps, texture naming conventions, or resolution guidelines.

Turn a Discord channel message into a tracked task by creating a thread, posting a parent-channel notice with the thread link (visibility), and seeding the thread with a structured template. Use when a user says “create a task/thread for this”, “start tracking this requirement/bug”, or when the workflow requires one-task–one-thread discipline in Discord.

Problem-solving strategies for rings in abstract algebra

Search Tool Hierarchy

Provides comprehensive guidance for Bootstrap Vue 3 component library including Bootstrap components, grid system, utilities, and Vue 3 integration. Use when the user asks about Bootstrap Vue 3, needs to use Bootstrap components in Vue 3, or implement responsive layouts.

Provides comprehensive guidance for Tencent Cloud COS including bucket creation, object storage, access control, and COS management. Use when the user asks about Tencent Cloud COS, needs to store files in COS, configure COS buckets, or work with Tencent Cloud storage.

Provides comprehensive guidance for hexagonal architecture including ports and adapters, domain isolation, and dependency inversion. Use when the user asks about hexagonal architecture, needs to implement ports and adapters pattern, or structure applications with hexagonal architecture.

A comprehensive skill for uni-ad monetization in uni-app. Use this skill to integrate ads, configure ad types, handle ad events, and follow official uni-ad documentation.

A comprehensive skill for uni-app native app offline packaging. Use this skill to package uni-app as native Android/iOS apps, configure native app settings, handle signing, and follow official offline packaging documentation.

Guidance for Vite using the official Guide, Config Reference, and Plugins pages. Use when the user needs Vite setup, configuration, or plugin selection details.

Supports evaluation of translational potential for target-based drug research outcomes and preparation of asset business plans (BP). Applicable to organizing targets/drug findings from universities, hospitals, or research institutions into translational maturity, intellectual property, competitive landscape, partner matching, and asset BPs.

JavaScript runtime performance patterns for hot paths. Use when optimizing loops, lookups, DOM operations, caching, or data structure choices. Framework-agnostic patterns that apply to React, Vue, vanilla JS, and any web application. Focus on measurable improvements in real-world hot paths.

Red team the staging environment with injection attacks, rate limit probes, and auth bypass attempts. Use before major releases or when adding new user-input surfaces.

Effect-TS patterns for getting started. Use when working with getting started material in Effect-TS applications.