fashishir

Facode — Design skill for Claude Code

Design community

How to install Facode

This entry records only its repository, not the path inside it, so there is no exact command to give. Open fashishir/facode and copy the folder into ~/.claude/skills/, or the file into ~/.claude/agents/.

What Facode does

FA Code is a Claude Code–style CLI coding agent with [REA (Reverse Engineering Anything)].it's an autonomous terminal agent for software engineering and binary analysis: streaming TUI, tool permission prompts, plan mode, sessions, context compaction, skills, subagents — and 200+ reverse-engineering tools exposed directly to the model.

Alternatives in Design

Extract — Design System & Product Reverse-Engineering - Extract tokens, components, architecture, and PRDs from codebase 2.8k ★
Deepclaude — Use Claude Code's autonomous agent loop with DeepSeek V4 Pro, OpenRouter, or any Anthropic-compatible backend 2.3k ★
Clui Cc — A lightweight, transparent desktop overlay for Claude Code on macOS 956 ★

README

FA Code (facode-cli)

**FA Code is a Claude Code–style CLI coding agent with [REA (Reverse Engineering Anything)](https://github.com/morluto/rea-agents) built in.**

It's an autonomous terminal agent for software engineering and binary analysis: streaming TUI, tool permission prompts, plan mode, sessions, context compaction, skills, subagents — and 200+ reverse-engineering tools (decompilers, disassemblers, APK/PE/Electron/JS analysis) exposed directly to the model.

Install

Requires **Node.js ≥ 20**.

npm install -g facode-cli

Then run from any directory:

facode

Quick start

# Interactive TUI (Claude Code style)
facode

# One-shot prompt
facode "explain the architecture of this repo"
facode -p "what is 2+2"

# Minimal UI (no Ink, plain text streaming)
facode --bare

# Pick a model / provider for this run (not persisted)
facode --provider deepseek --model deepseek-chat

⚠️ The prompt is a **positional argument**, not a flag: `facode "your prompt"`. Use `-p`/`--prompt` for an explicit one-shot run.

Configuration

FA Code needs an **OpenAI-compatible** LLM endpoint. Configuration layers (later wins):

Built-in presets: openai, deepseek, pi, anthropic, ollama
User config: ~/.facode/config.json — applies everywhere
Project config: .facode/config.json — applies in this project
Environment: FACODE_API_KEY, FACODE_BASE_URL, FACODE_MODEL, FACODE_PROVIDER (OPENAI_API_KEY / OPENAI_BASE_URL are honored by the openai preset)

Providers & models

# Register a custom provider (saved to ~/.facode/config.json — works everywhere)
facode provider add myproxy --base-url https://my.proxy/v1 --api-key sk-... --model my-model

# Or reference an environment variable instead of storing the key on disk
facode provider add myproxy --base-url https://my.proxy/v1 --api-key-env MY_PROXY_KEY

# Bare `facode provider add` (in a t