Browse Skills

DingTalk Documents accesses DingTalk knowledge spaces, folders, and documents through the DingTalk document MCP. Use it when a user selects a DingTalk source and you need to traverse or read its content.

Hello Wework calls the bundled hello_wework MCP tool with a user's name and returns the greeting. Use it to demonstrate a complete Wework plugin flow.

Image Generation calls generate_image when a user wants to create, draw, or revise an image. It supports text-to-image prompts and reference images from user attachments or URLs.

Prompt Optimization lets you view and revise an agent’s system prompt and member prompts. Use it when you want to change the agent’s behavior, focus, or response style.

Skill Creator helps create new skills, refine existing ones, and measure how well they trigger and perform. Use it when you want to draft, edit, evaluate, benchmark, or optimize a skill and its description.

Video Generation creates videos when a user asks for one. It accepts text, image, video, and audio references and uses the selected model to generate the result.

Wegent Knowledge manages knowledge bases and documents with tools to list, create, update, read, and search content using RAG retrieval. Use it when you need to organize or query knowledge programmatically.

Agent Evals builds repeatable tests for AI agent behavior using golden datasets, tool-call checks, safety prompts, and judge-based scoring. Use it to validate prompt changes, catch regressions, and gate deployments on quality.

Agent Observability instruments AI agents with logs, traces, metrics, token usage, latency, and cost telemetry. Use it to debug reliability issues, set SLOs, and monitor agent or LLM workloads.

AI Pipeline Orchestration builds reliable workflows for document ingestion, batch inference, model training, and RAG indexing. Use it to schedule recurring AI jobs and manage dependencies in production pipelines.

LLM Caching implements exact-match, semantic, and provider-side prompt caching for LLM calls. Use it when repeated or similar prompts need lower API cost, faster responses, or higher throughput.

LLM Cost Optimization reduces API and infrastructure spend with model right-sizing, caching, batching, prompt compression, quantization, and self-hosting. Use it when you need to track AI costs, set budgets, or route requests by cost.