Create and orchestrate GenAIScript scripts for LLM prompts and AI workflows.
npx skills add https://github.com/markpitt/claude-skills --skill genaiscript
Please help me install this Agent Skill. Skill: genaiscript Source: https://github.com/markpitt/claude-skills/tree/main/skills/genaiscript Command: npx skills add https://github.com/markpitt/claude-skills --skill genaiscript
This Skill provides a structured gateway to GenAIScript documentation and patterns for building, debugging, and optimizing prompts, prompts‑as‑code, and AI workflows.
Load core concepts to learn the syntax of script(), def(), defData(), and defTool(), then explore examples to accelerate real workflows.
High-intent search queries and answers about installing and using this skill.
GenAIScript enables prompts-as-code by letting you write scripts using script() and def() functions to structure LLM interactions. You define prompt templates with $-syntax, register tools with defTool(), and orchestrate multi-step AI workflows in a single declarative script that runs end-to-end.
Yes. GenAIScript supports file processing for PDF, DOCX, and CSV formats through defData() and content inclusion APIs. You can load and parse these files within scripts to feed structured data into LLM prompts and automate document-based AI tasks.
Use defSchema() to declare structured data formats and defTool() to register external functions that LLMs can invoke. This lets you create type-safe tool integrations and enforce output structure, enabling agents to interact with APIs and services within orchestrated workflows.
GenAIScript uses design-pattern-driven workflows where you chain script() calls, define tool dependencies with defTool(), and manage state across steps. You compose scripts with schema validation, file outputs via defFileOutput(), and MCP support to create scalable, modular agent automation.
Yes. GenAIScript includes MCP support built into its architecture, allowing you to connect standardized protocol-based tools and services directly into your prompts-as-code workflows without custom wrapper code.
GenAIScript workflows account for LLM latency, token costs, and sequential step execution. Design patterns emphasize batching file operations, caching schema definitions, and parallelizing independent tool calls to optimize throughput in multi-step automation.
hxk622Convert PDF, DOCX, PPTX, XLSX, images, audio, HTML, CSV, JSON, XML, ZIP, EPubs, and YouTube URLs into Markdown.
imehrLog AI executions and learn from feedback using JSONL logs and JSON recommendations.
derekhu0002Automate MCP server development with Node/TypeScript SDKs and transport patterns.
zmreImprove prompts for AI agents using prompt engineering standards.
