All prompts

Performs a rapid security assessment of open-source projects, identifying common vulnerability patterns and supply chain risks

Analyzes the composition of your LLM application prompts, identifies token waste points, and provides compression and optimization suggestions to fit more effective information within a limited context window.

Automatically decompose complex tasks into subtasks, assign them to multiple AI agents for parallel execution, and finally aggregate the results. Suitable for large-scale tasks such as research reports and code projects.

Generate high-quality plugin/skill development documentation for AI programming assistants (e.g., Claude Code, Cursor), including architecture design, interface specifications, and example code.

Helps developers analyze and debug the tool call chains of AI Agents, identifying failure nodes and optimizing call strategies.

Automatically generate professional contribution guides for open-source projects, including coding standards, PR processes, and issue templates.

Simulate professional financial researchers to conduct multi-dimensional deep analysis of companies or industries, outputting structured investment research reports.

Design local multi-model dynamic loading/unloading solutions to achieve on-demand switching between different LLM models, optimizing VRAM usage and latency.

Designs terminal-based AI agent automation workflows, including task orchestration, error handling, and human-machine collaboration.

Convert complex document images (tables, forms, handwriting) into structured data, supporting multiple output formats.

Helps individuals and small businesses set up localized AI bookkeeping systems to automatically classify invoices and transaction records.

Helps developers evaluate and select toolchain components (MCP servers, file search, memory systems, etc.) for AI Agent development, outputting integration plans and architectural advice.

Intelligently proofread OCR results, fixing recognition errors and restoring format structures.

Guides developers in using the Compound Engineering method to build multi-Agent systems, including task decomposition, context management, tool orchestration, and quality feedback loops

Helps you plan and design local visual-language model (VLM) application solutions on Mac, including model selection, performance optimization, and deployment strategies.

Guides users in deploying, fine-tuning, and evaluating Vision-Language Models (VLMs) in local environments (especially Apple Silicon Macs), including performance optimization and benchmark comparisons.

Help diagnose bottlenecks, failure points, and context loss issues in AI Agent operations, generating observability plans.

Input a GitHub project link, and AI helps you quickly evaluate the project's quality and development potential from three dimensions: technology, community, and commercialization.

Quickly designs the technical architecture and interaction flow of voice AI applications, covering core modules such as ASR, TTS, and dialog management

Design multimodal data pipelines for AI projects to process images, audio, video, and structured data, including ETL processes and performance optimization.

Helps you analyze voice AI application scenarios, recommends suitable open-source voice models and tech stacks, and generates a complete technology selection report.

Paste any GitHub README content, and AI will score it across 10 dimensions and provide specific optimization suggestions.

Enables AI to automatically recognize receipt/invoice content, extract key fields, and classify/archive them, suitable for smart bookkeeping scenarios for freelancers and small teams

Batch-generate short video topics, titles, and hook openings based on platform trends and user personas, suitable for Douyin, Xiaohongshu, YouTube Shorts.

Automatically rewrite an article into multiple versions suitable for different platforms (Xiaohongshu, Twitter, LinkedIn, WeChat public accounts).

Help analyze and compress context in LLM applications to reduce token consumption while maintaining information integrity.

Systematically compares the workflows, pros/cons, and best-use scenarios of different AI coding assistants.

Helps developers plan and manage AI Agent tool package dependencies, version compatibility, and deployment strategies, similar to npm/pip but for the Agent ecosystem.

Designs quantitative trading strategies based on multi-agent collaboration, including role division for researchers, risk control, and execution, outputting a complete backtesting plan.

Starts from product requirements to help you choose the right voice AI tech stack (TTS/STT/Voice Clone) and outputs a complete solution.

Input a description of abnormal behavior encountered by your AI Agent to automatically generate a structured troubleshooting decision tree, diagnosing from symptoms to root causes layer by layer.

Decompose a complex business requirement into role definitions, communication protocols, and collaboration flowcharts for multiple AI Agents.

Performs deep performance reviews on Rust code, providing specific optimization suggestions and rewrite examples covering memory allocation, concurrency models, compiler optimizations, and zero-copy strategies.

Helps you design long- and short-term memory architectures for AI Agents, including session state persistence, vector retrieval, and summarization compression strategies.

Describe your business needs, and AI helps you generate a one-page system architecture design, including component diagrams, technology selection, data flow, and deployment plans.

Performs anomaly detection analysis on time series data, generating visual reports and alert recommendations. Suitable for business monitoring, IoT sensors, and financial data scenarios.

Compare mainstream multi-agent orchestration frameworks (oh-my-claudecode, CrewAI, AutoGen, LangGraph, etc.) and provide selection advice and migration plans.

Design a unified AI model gateway to implement multi-vendor routing, failover, cost control, and observability.

Help developers design the most suitable memory layer solutions for AI Agents, including short-term/long-term memory, vector storage, and state synchronization technology selection.

Helps design reinforcement learning training schemes for AI Agents, including reward function design, trajectory sampling strategies, and automatic prompt optimization, applicable to any Agent framework.

Conduct a comprehensive security audit of Bash scripts, identifying injection vulnerabilities, permission issues, and unsafe practices, and provide a hardened secure version. Suitable for security reviews of Agent automation scripts.

Automatically generate integration test cases for CLI Agent toolchains, covering normal flows, edge cases, and exception scenarios to ensure reliable tool execution across various environments.

Diagnoses reasons for poor retrieval performance in your RAG system and provides specific optimization plans.

Designs a 0-to-1 cold start growth strategy for new products, including user acquisition, retention, and viral propagation plans.

Recommends the best time series forecasting model based on data characteristics and provides tuning strategies and evaluation plans.

A professional context engineering consultant that helps you maximize LLM output quality within limited token budgets, suitable for long document processing, multi-turn conversations, and RAG scenarios.

Breaks down penetration testing targets into automated sub-task chains, supporting the full process from information gathering and vulnerability scanning to exploitation and report generation.

Designs multi-agent collaboration systems for robot hardware platforms, supporting natural language control of humanoid robots, quadrupeds, and drones.