Agent Skills · AI Agent
Curated skills for Claude, Cursor and other AI agents. Copy one prompt to install.
Organize provided prompts into clear names, tags, duplicate decisions, and reuse notes.
Helps individuals and small teams turn AI subscriptions, API usage, and recurring model tasks into a cost-control plan with routing rules, budget guardrails, and review checklists.
Creates, restores, inspects, exports, and imports working-context snapshots for AI agents.
An agent workflow to identify skill gaps, search and audit sources, install packages, generate skills, and iterate updates.
A Python/FastAPI and React quant trading agent with Sina quotes, 10-dimension scoring, risk controls, order execution, and self-learning review.
Switch between multiple personas so conversations show distinct tones, personalities, and interaction styles.
An AI multi-agent collaboration system based on the Three Excellencies and Nine Ministers, with tag routing, model tiers, task queues, agent memory, knowledge retrieval, and audit controls.
A meta-skill for AI agents that generates, merges, scores, and syncs publishable skill packages with monetization plans.
A standardized three-step AI workflow from requirements to runnable code.
Monitor WeChat, Zhihu, and GitHub to aggregate yesterday's deep recommendation-ads and LLM articles plus open-source trends into a structured brief.
A behavior training pack for AI agents that converts mistakes into reusable rules through four-step retrospectives, layered memory, and verification loops.
Switch the current session between DeepSeek V4 Pro and V4 Flash by intent, with a cost note.