outlines
Guarantee valid JSON/XML/code structure during generation, use Pydantic models for type-safe outputs, support local models (Transformers, vLLM), and maximize inference speed with Outlines - dottxt.ai's structured generation library
sglang
Fast structured generation and serving for LLMs with RadixAttention prefix caching. Use for JSON/regex outputs, constrained decoding, agentic workflows with tool calls, or when you need 5× faster inference than vLLM with prefix sharing. Powers 300,000+ GPUs at xAI, AMD, NVIDIA, and LinkedIn.
guidance
Control LLM output with regex and grammars, guarantee valid JSON/XML/code generation, enforce structured formats, and build multi-step workflows with Guidance - Microsoft Research's constrained generation framework
instructor
Extract structured data from LLM responses with Pydantic validation, retry failed extractions automatically, parse complex JSON with type safety, and stream partial results with Instructor - battle-tested structured output library
receipt-scanner
Extract vendor, date, items, amounts, and total from receipt images using OCR and pattern matching with structured JSON output.
ocr-document-processor
Extract text from images and scanned PDFs using OCR. Supports 100+ languages, table detection, structured output (markdown/JSON), and batch processing.
meeting-notes-formatter
Convert raw meeting notes to structured markdown or PDF with automatic section detection, action items extraction, and attendee parsing.
review-format
Output in the specified format when a review of specific markdown is requested.
prompt-refinement
Transform prompts into structured TCRO format with phase-specific clarification. Automatically invoked by /ai-eng/research, /ai-eng/plan, /ai-eng/work, and /ai-eng/specify commands. Use when refining vague prompts, structuring requirements, or enhancing user input quality before execution.
context-setter
Generate comprehensive context-setting markdown files for coding tasks by intelligently exploring codebase and structuring work with ROLE, CONSTRAINTS, STYLE, file reading order, and structured tasks. Use proactively when the user is starting a complex dev task requiring multiple steps or file reads.
analyzing-research-papers
Expert methodology for analyzing and summarizing research papers, extracting key contributions, methodological details, and contextualizing findings. Use when reading papers from PDFs, DOIs, or URLs to create structured summaries for researchers.
brainstorming
Use when creating or developing anything, before writing code or implementation plans - refines rough ideas into fully-formed designs through structured Socratic questioning, alternative exploration, and incremental validation. Optimized for git worktree workflows and Claude CLI agent patterns.
youtube-to-markdown
Use when user asks YouTube video extraction, get, fetch, transcripts, subtitles, or captions. Writes video details and transcription into structured markdown file.
perplexity-research
Use when conducting web research with Perplexity API - covers authentication, search strategies, structured output generation, and integration patterns for Claude-interpretable results
pydantic-ai-agent-creation
Create PydanticAI agents with type-safe dependencies, structured outputs, and proper configuration. Use when building AI agents, creating chat systems, or integrating LLMs with Pydantic validation.
research-methodology
Activate when conducting topic research with complexity assessment, thinking mode selection, and comprehensive documentation generation
invoking-gemini
Invokes Google Gemini models for structured outputs, multi-modal tasks, and Google-specific features. Use when users request Gemini, structured JSON output, Google API integration, or cost-effective parallel processing.
