Awesome AI Agents › AI Developer Toolkit

QingYunA/answer-me-with-html

⭐ 6091 JavaScript added to this list on 2026-11-09 repository created 2026-10-02

Answer me with HTML is a skill for coding and chat agents (Claude Code, Codex, Cursor, OpenCode, Pi, DeepSeek Harness) that turns a plain-language answer into a readable, offline, single-file HTML page rather than a long block of terminal text. The agent still writes only a short Markdown draft of the actual content; a bundled Node.js CLI shipped inside the skill then generates the surrounding page -- SVG diagrams, CSS and HTML structure -- in roughly 50 milliseconds. The authors measured that hand-written HTML answers spend most of their tokens on presentation rather than content (47% on SVG diagram coordinates and paths, 15% on CSS, 17% on HTML tags, only 21% on actual text, nearly 4,900 tokens on average across nine sample pages); offloading presentation to the CLI cut the model's own output for the same answers to about one eighth as much, and roughly 2.6x faster end to end, with explainer videos saving proportionally more. Example outputs include a sequence/state diagram and flag table for explaining the TCP handshake, a folder tree and call graph for mapping a repository's modules, a pros/cons comparison table with a verdict for choosing between two technologies, and a timeline for historical questions; trivial one-line questions still get a one-line text answer rather than a page. Generated pages are saved locally and include a UI for switching theme and light/dark mode. Installation is handled by letting the user's own agent follow an INSTALL.md, through a Claude Code plugin marketplace, a DeepSeek Harness plugin command, or a cross-agent installer supporting more than 70 agent targets; an "always-on mode" can be enabled so every answer, not only complex ones, gets a page. Requires Node.js 20+, written primarily in JavaScript.

https://github.com/QingYunA/answer-me-with-html

agent-skillhtml-generationdocumentationclaude-code-plugindeveloper-toolsjavascript

Also in AI Developer Toolkit

Leonxlnx/taste-skill

Taste Skill provides portable AI Agent Skills designed to enhance the quality of AI-generated frontends and image-based design references, aiming to eliminate generic UI outputs.

unslothai/unsloth

Unsloth is a toolkit for fine-tuning and reinforcement learning of large language models, enabling faster training with significantly reduced VRAM usage across various advanced models and tasks.

virgiliojr94/book-to-skill

A command-line tool that converts technical books, document folders or globs into structured agent skills in the open SKILL.md format, so coding agents load only the chapter relevant to a question.

reworkd/AgentGPT

AgentGPT is an open-source platform that allows users to assemble, configure, and deploy autonomous AI agents in a web browser to pursue custom goals by generating and executing tasks independently.

google/adk-python

An open-source Python toolkit by Google for building, evaluating, and deploying sophisticated AI agents with flexibility, modularity, and integration within the Google ecosystem.

can1357/oh-my-pi

Oh My Pi is an AI coding agent for the terminal featuring hash-anchored edits, an optimized tool harness, LSP integration, and support for Python, browsers, and subagents.

millionco/react-doctor

React Doctor is an intelligent agent designed to diagnose and fix issues within React codebases, leveraging AI to improve code quality and developer efficiency.

max-sixty/worktrunk

Worktrunk is a CLI for Git worktree management, streamlining parallel AI agent workflows and simplifying common Git worktree tasks with advanced automation features.