All prompts

Input the product name and core selling points to automatically generate launch copy suitable for both Twitter/X and Xiaohongshu, including hooks, feature highlights, CTAs, and tags.

Design a long-term memory system for an AI assistant based on knowledge graphs, supporting automatic construction and retrieval of context from emails, meeting notes, and other sources.

Design a collaborative management scheme for multiple AI coding agents working in parallel, including task assignment, progress tracking, and skill accumulation.

Input a natural language description to generate CAD modeling instructions, parametric design schemes, and 3D model structure descriptions, suitable for rapid prototyping and engineering modeling scenarios.

Designed isolated execution strategies for multi-agent parallel coding scenarios, including task allocation, branch management, conflict detection, and automatic merging schemes, suitable for autonomous coding platforms such as Symphony and OpenHands.

Design a personal AI super assistant system architecture with long-term memory, multi-platform integration, and active perception capabilities

Design solutions for remotely managing AI encoding agents from mobile phones, including session synchronization, voice control, notification, and approval workflows

For visual AI or NLP datasets, systematically evaluate annotation quality, distribution bias, duplicate samples, and other issues, and generate actionable cleaning solutions

Based on the user's hardware configuration (GPU/CPU/memory), it automatically generates the optimal local LLM inference service deployment script, supporting llama.cpp, vLLM, Ollama solution selection and parameter optimization.

Simulate the thinking styles of multiple investment masters to conduct multi-angle analysis of specified stocks and generate comprehensive investment advice.

Convert natural language descriptions into precise AI video editing instructions, supporting scene decomposition, transition design, and style consistency control.

Design a full-stack observability solution for AI Agent applications, covering LLM call tracing, tool execution monitoring, cost analysis, and anomaly detection dashboards.

Describe your AI Agent workflow to automatically generate Mermaid flowchart code, which can be directly pasted into documents for rendering.

Design a secure sandbox execution environment for AI Agents based on WebAssembly, including resource limits, permission models, and escape prevention.

Design an Agent memory system based on a temporal knowledge graph, supporting fact version tracking, incremental updates, and hybrid retrieval.

Design a rigorous A/B test experiment plan for your LLM application, including sample size calculation, evaluation metric definition, statistical significance testing, and result visualization code.

Input your Agent tool list and common error scenarios to automatically generate a complete error handling decision tree, including retry strategies, fallback plans, and user notification templates.

Input model name and deployment environment to automatically generate a comparative evaluation of inference acceleration technologies, including quantization, distillation, and speculative decoding.

Have the AI act as a quantitative trading analyst, fusing multiple data sources (news, social media, K-lines, on-chain data) to generate structured trading signals, including entry points, stop-loss/take-profit levels, and confidence scores.

Generate a complete performance benchmarking plan for an AI Agent's sandbox execution environment, including test cases and evaluation criteria for key metrics such as cold start latency, memory isolation, and concurrent execution.

Generate a robust task execution plan for a browser automation AI Agent, including element location strategies, wait conditions, error recovery mechanisms, and human-in-the-loop breakpoint designs.

Design memory compression strategies for AI coding agents to automatically capture key decisions and context, restoring continuity in new sessions.

Input a topic and style to automatically generate a complete short video script, including storyboard descriptions, dialogue, duration, and visual cues, suitable for Douyin/Xiaohongshu/YouTube Shorts.

Generates a complete self-hosted deployment plan for open-source AI clients (such as Thunderbolt, Open WebUI, LibreChat, etc.) based on user hardware conditions and requirements, including model selection, Docker configuration, and security hardening.

Generate a semantic search indexing strategy for your code repository, enabling AI coding assistants to precisely locate any code snippet.

Compresses lengthy AI coding session histories into structured memory files and accurately restores context in new sessions, significantly reducing token consumption while preserving key decisions and code changes.

Transform vague project requirements into structured Agent team task assignment schemes, including role definitions, task breakdowns, dependencies, and acceptance criteria. Suitable for multi-agent collaboration tools such as Squad and Claude Agent Teams.

Optimized strategies for the context window of encoding agents, achieving a 98% token compression rate through sandboxing tool output. Efficiency optimization for AI programming tools such as Claude Code, Codex, and Cursor.

Generate complete product design documentation for multimodal AI Agent desktop applications, covering core capabilities such as GUI operations, browser automation, and MCP tool integration.

Help developers quickly design tool discovery, MCP Server integration, and database connection solutions for AI Agents, generating complete configuration and code templates.

Help you generate high-quality synthetic training data from scratch, supporting custom domains, formats, and quality standards, suitable for fine-tuning and evaluation scenarios.

Generate selection recommendations and setup steps for zero-code or low-code AI Agent frameworks based on your needs.

Design asynchronous task orchestration schemes for multi-step AI Agents, including retry, timeout, callback, and state machine design.

Generate code change impact scope reports based on semantic analysis rather than text diffs, identifying potential risks and modules requiring regression testing.

Design fine-grained sandbox security policies for AI Agent applications, including file system, network, and process permission controls, suitable for WebAssembly isolation and containerized deployment scenarios.

Design an end-to-end financial data prediction and analysis solution based on the latest financial time series foundation models (such as Kronos).

Generate secure sandbox configuration policies for AI agents executing multiple tools, including sandbox isolation and permission control for interfaces such as OpenAPI, MCP, and GraphQL.

Design fine-tuning and deployment schemes tailored to specific financial scenarios based on time series foundation models like TimesFM.

Input your agent's tool list and common error scenarios to automatically generate a comprehensive error-handling decision tree, including retry strategies, fallback plans, and user notification templates.

Describe your document types and scale to generate a complete local semantic search system deployment plan, including embedding model selection, vector database configuration, and search API code.

Design a rigorous A/B test experiment plan for your LLM application, including sample size calculation, evaluation metric definition, statistical significance testing, and result visualization code.

Build a financial time series analysis framework using AI, combining foundation models for market trend prediction and risk assessment.

Input project requirements to automatically generate a multi-agent collaboration task board, including task assignment, dependencies, and priority sorting.

One-click generation of the complete technical architecture for voice AI applications, covering STT, TTS, VAD, dialogue management, and multi-channel integration.

Use AI to automatically analyze scattered notes in an Obsidian vault, generating reasonable bi-directional links, tag systems, and MOCs (Maps of Content), transforming the knowledge base from a pile of information into a network of knowledge.

Input exported chat history into the AI to automatically generate social relationship insights, topic trends, sentiment analysis, and annual review reports.

Describe requirements in natural language; AI automatically breaks them down into subtasks for multiple coding agents and generates a progress tracking dashboard.

Analyze whether text or images contain AI-generated watermarks (such as SynthID) and provide a detection report with confidence assessment.