quinteroac

Install ComfyUI Assistant as a ComfyUI custom node. Use when performing or troubleshooting installation (clone, Python deps, frontend build, restart).

For complex user requests — evaluate, investigate, ask questions, propose a plan, accept modifications, then execute. Use when the request is multi-step, ambiguous, or high-impact.

Workflow execution and complete workflow generation tools (executeWorkflow, applyWorkflowJson). Use when the user wants to run a workflow or build a complete workflow from a description.

Python backend modules, chat request lifecycle, SSE format, and API endpoints

System prompt assembly from system_context, user_context, and environment sources

Generate a Product Requirements Document (PRD) for a new feature. Use when planning a feature, starting a new project, or when asked to create a PRD. Triggers on: create a prd, write prd for, plan this feature, requirements for, spec out.

When the user asks to create or design a skill, follow a clear process: clarify intent, draft name/description/instructions, then call createSkill. Optionally offer to create a skill when you automate a procedure they might want to reuse.

When the user asks to organize a workflow, read frontend JSON, reorganize for readability, apply in a new tab, and always validate with getWorkflowInfo. Run automatically without asking.

Implement React table components with reflow-free text measurement using @chenglou/pretext. Use this skill when the user asks to build a table with dynamic row heights, virtualized scroll, resizable columns, inline editing, drag-to-reorder, or any table feature where accurate pre-render text measurement matters. Triggers on: "build a table", "add a table component", "virtualized table", "table with dynamic row heights", "resizable columns", "pretext table", "use pretext for a table", "table without reflow".

Drive interactive approval of the prototype by updating project context and roadmap artifacts after a completed iteration.

Validate the current iteration's PRD against the implemented code via the audit prototype skill.

Creates or updates .agents/PROJECT_CONTEXT.md with project conventions, tech stack, code standards, testing strategy, and product architecture. Triggered by: nvst create project-context.

Implements a single user story from the PRD: writes code and tests, follows project conventions. Invoked by: nvst create prototype.

Gathers the requirement from the user and produces it_{iteration}_product-requirement-document.md. Triggered by: nvst define requirement.

Fixes one issue from the iteration issues list by reproducing, diagnosing, and resolving it. Invoked by: bun nvst fix issue.

Create distinctive, production-grade frontend interfaces with high design quality. Use this skill when the user asks to build web components, pages, artifacts, posters, or applications. Generates creative, polished code that avoids generic AI aesthetics.

Fixes one issue from the iteration issues list by reproducing, diagnosing, and resolving it. Invoked by: bun nvst fix issue.

Refines .agents/PROJECT_CONTEXT.md via editor mode or challenge mode. Challenge mode validates the document against the actual codebase and detects compliance issues. Triggered by: nvst refine project-context.

Leads a structured ideation interview to surface feature/project ideas and update ROADMAP.md with new candidates.

Creates or updates .agents/PROJECT_CONTEXT.md with project conventions, tech stack, code standards, testing strategy, and product architecture. Triggered by: nvst create project-context.

Drive interactive approval of the prototype by updating project context and roadmap artifacts after a completed iteration.

Load the refactor plan from it_{iteration}_audit.json and invoke the agent to apply code changes.

Validate the current iteration's PRD against the implemented code via the audit prototype skill.

Implements a single user story from the PRD: writes code and tests, follows project conventions. Invoked by: nvst create prototype.

Helps entrepreneurs discover viable markets quickly through a guided profiling interview, then recommends 3–5 tailored market opportunities in a lean, actionable format. Use when the user says "help me find a market", "market opportunities for me", "find me a business niche", "I want to start a business but don't know where", "what markets can I explore", "looking for business opportunities", or asks for entrepreneurial direction without a clear idea yet.

Helps developers choose their next programming project through a guided interview covering motivation, niche, pain points, and format, then researches the web to surface existing ready-to-use solutions and propose 4 concrete ideas with competitive analysis, differentiation factors, and effort estimates. Use when the user says "help me choose my next project", "I don't know what to build next", "what should I build", "help me pick a project", "suggest a programming project for me", "ayúdame a elegir mi próximo proyecto", or asks for project ideas without a clear direction.

Scaffold and create a new skill for this repository. Use when the user says "create a new skill", "add a skill for...", "scaffold a skill", "I want to build a skill that...", "make a skill that does X", "help me add a capability", or whenever they want to add a new capability to this skills pack. Also use when improving the structure of an existing skill.

Creates an agent skill from the official documentation of any tool, library, or API. Use when the user shares a documentation link and wants to turn it into an installable skill, when they say "I want a skill for X", "create a skill from this doc", "make a skill from [URL]", "we need our agents to work with X", "turn this documentation into a skill", or when they want their agents to understand a specific tool — even if they don't explicitly use the word "skill".

Skill for building documentation websites with Docusaurus — helps agents scaffold sites, configure plugins/themes, write content in MDX, and deploy to platforms like Vercel, Netlify, or GitHub Pages.

Generates creative and artistic text layout code using @chenglou/pretext — for kinetic typography, text-along-a-path, generative type art, text masking, shaped text containers, wavy/curved text, typographic posters, and custom Canvas/SVG text rendering. Use when user says "creative text layout with pretext", "text art with pretext", "kinetic typography", "text on a path", "generative text", "typographic art", "shaped text", "wavy text", "text poster", "artistic text canvas", "pretext creative", or when they want visually expressive text that goes beyond simple paragraph wrapping.

API reference for @chenglou/pretext — a pure JS/TS library for measuring and laying out multiline text without DOM reflows. Use when you need to look up a pretext function signature, understand the two-phase prepare/layout model, find the right function for a use case (height-only vs line content vs variable-width), check PrepareOptions or LayoutCursor types, or orient yourself before writing code that uses pretext. Also use when user says "pretext API", "pretext docs", "pretext types", "how does pretext work", "pretext functions", or "what does prepare/layout/layoutWithLines/walkLineRanges do".

Guides implementation of @chenglou/pretext in a real project — asks a few targeted questions about rendering target, data shape, and performance context, then generates correct, complete integration code. Use when user says "help me integrate pretext", "how do I use pretext to measure text", "I need to calculate text height", "text layout with pretext", "pretext in React", "pretext for Canvas", "pretext for virtualization", "pretext SSR", "shrink-wrap a container to text", "multiline text height calculation", "pretext without DOM", or when they need working pretext code for a specific context.

Drive interactive approval of the prototype by updating project context and roadmap artifacts after a completed iteration.

Validate the current iteration's PRD against the implemented code via the audit prototype skill.

Gathers the requirement from the user and produces it_{iteration}_product-requirement-document.md. Triggered by: nvst define requirement.

Creates or updates .agents/PROJECT_CONTEXT.md with project conventions, tech stack, code standards, testing strategy, and product architecture. Triggered by: nvst create project-context.

Implements a single user story from the PRD: writes code and tests, follows project conventions. Invoked by: nvst create prototype.

Load the refactor plan from it_{iteration}_audit.json and invoke the agent to apply code changes.

Updates an existing product requirement document based on user feedback. Triggered by: nvst refine requirement.

Read and analyze ComfyUI workflow JSON files. Expands subgraph nodes recursively and exposes every node with its type, parameters, connections, and execution state. Designed for pipeline analysis — finding which models are needed, tracing execution flow, and detecting bypassed nodes.

Project conventions and complete public API reference for comfy_diffusion — covers coding patterns, data formats, and every module signature.

Personalized brainstorming agent that generates sellable utility ideas (CLI, TUI, GUI, agent tools) tailored to the user's stack and niche, with persistent memory of past suggestions. Triggered by: /spark.

Generate an SVG logo from a text prompt using Zikon's pipeline (prompt → PNG → SVG). Returns a JSON object with paths and inline SVG markup.