Agent Skills · AI Agent
Curated skills for Claude, Cursor and other AI agents. Copy one prompt to install.
A universal task skill for large models and AI agents.
Reviews and fixes AI Agent instructions, tools, workflows, and scoring with a product-manager perspective.
A Kunlun ecosystem skill package for AI Agents.
Provides lightweight lifecycle management for local skill files, deciding whether to retain, improve, merge, or create reusable skills based on session evidence.
Install a complete Feishu skill pack covering docs, messages, calendar, tasks, Bitable, Drive, and Wiki.
Generates an SLG gaming industry daily briefing, covering AI news, new titles, competitor trends, tests, Steam releases, and platform banners.
A zero-dependency AI Agent workflow orchestrator that supports natural-language DAG generation, data nodes, conditional routing, parallel execution, and skill calls with SQL allowlists, path validation, and resource limits.
For beginners and business users, it clarifies fuzzy workbench requirements into a core workflow and generates a complete prompt for a workbench builder agent.
Detects and configures R, Stata, SPSS, and other statistical tools to connect legacy scripts, data conversion, and batch runs into AI workflows.
A hallucination-control skill for AI agents that enforces fact checking, source verification, confidence labeling, and two-pass self-review to reduce fabrication and over-compliance.
Keeps the agent thinking when idle and proactively continuing conversations from workspace knowledge and user ideas.
A workflow for creating and iterating Claude skills: capture intent, draft SKILL.md, run test prompts, review outputs, optimize trigger descriptions, and package the final skill.