Showcase
Amplifier TUI
A terminal interface that makes AI agent work visible, navigable, and beautiful.
Built with Textual + Amplifier  |  78+ commands  |  11 themes  |  922 tests
Active
The Interface
Your AI conversation, in the terminal
Amplifier TUI …/ANext/amplifier-tui How do I set up a Python virtual environment? ▶ Thinking: The user is asking about Python virtual environments... ▶ Tool: bash Virtual Environments Create one with the built-in venv module: ────────────────────────────────────────────────────────────────────────────────────────────── python3 -m venv .venv source .venv/bin/activate ────────────────────────────────────────────────────────────────────────────────────────────── Your prompt will show (.venv) when active. Install packages normally with pip install. Tip: Use uv for faster package management: ────────────────────────────────────────────────────────────────────────────────────────────── uv venv .venv uv pip install requests ────────────────────────────────────────────────────────────────────────────────────────────── What about conda? ▼ Thinking: Conda is a different ecosystem... Conda is a different ecosystem. I should compare it with venv/uv. Conda creates isolated environments with their own Python: ────────────────────────────────────────────────────────────────────────────────────────────── conda create -n myenv python=3.11 conda activate myenv▆▆ ────────────────────────────────────────────────────────────────────────────────────────────── ─────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────── ─────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────── No sessionR
Rich markdown rendering, collapsible thinking blocks, tool-use visualization, streaming responses — all in a terminal.
By the Numbers
78+
Slash commands
11
Color themes
922
Tests passing
14
Features built overnight
11 Built-In Themes
Your terminal, your aesthetic
Amplifier TUI …/ANext/amplifier-tui How do I set up a Python virtual environment? ▶ Thinking: The user is asking about Python virtual environments... ▶ Tool: bash Virtual Environments Create one with the built-in venv module: ───────────────────────────────────────────────────────────────────────────────────────────── python3 -m venv .venv source .venv/bin/activate ───────────────────────────────────────────────────────────────────────────────────────────── Your prompt will show (.venv) when active. Install packages normally with pip install. Tip: Use uv for faster package management: ───────────────────────────────────────────────────────────────────────────────────────────── uv venv .venv uv pip install requests ───────────────────────────────────────────────────────────────────────────────────────────── What about conda? ▼ Thinking: Conda is a different ecosystem... Conda is a different ecosystem. I should compare it with venv/uv. Conda creates isolated environments with their own Python: ───────────────────────────────────────────────────────────────────────────────────────────── conda create -n myenv python=3.11 conda activate myenv▆▆ ───────────────────────────────────────────────────────────────────────────────────────────── ─────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────── ─────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────── No sessionR
Catppuccin
Amplifier TUI …/ANext/amplifier-tui How do I set up a Python virtual environment? ▶ Thinking: The user is asking about Python virtual environments... ▶ Tool: bash Virtual Environments Create one with the built-in venv module: ───────────────────────────────────────────────────────────────────────────────────────────── python3 -m venv .venv source .venv/bin/activate ───────────────────────────────────────────────────────────────────────────────────────────── Your prompt will show (.venv) when active. Install packages normally with pip install. Tip: Use uv for faster package management: ───────────────────────────────────────────────────────────────────────────────────────────── uv venv .venv uv pip install requests ───────────────────────────────────────────────────────────────────────────────────────────── What about conda? ▼ Thinking: Conda is a different ecosystem... Conda is a different ecosystem. I should compare it with venv/uv. Conda creates isolated environments with their own Python: ───────────────────────────────────────────────────────────────────────────────────────────── conda create -n myenv python=3.11 conda activate myenv▆▆ ───────────────────────────────────────────────────────────────────────────────────────────── ─────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────── ─────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────── No sessionR
Nord
Amplifier TUI …/ANext/amplifier-tui How do I set up a Python virtual environment? ▶ Thinking: The user is asking about Python virtual environments... ▶ Tool: bash Virtual Environments Create one with the built-in venv module: ───────────────────────────────────────────────────────────────────────────────────────────── python3 -m venv .venv source .venv/bin/activate ───────────────────────────────────────────────────────────────────────────────────────────── Your prompt will show (.venv) when active. Install packages normally with pip install. Tip: Use uv for faster package management: ───────────────────────────────────────────────────────────────────────────────────────────── uv venv .venv uv pip install requests ───────────────────────────────────────────────────────────────────────────────────────────── What about conda? ▼ Thinking: Conda is a different ecosystem... Conda is a different ecosystem. I should compare it with venv/uv. Conda creates isolated environments with their own Python: ───────────────────────────────────────────────────────────────────────────────────────────── conda create -n myenv python=3.11 conda activate myenv▆▆ ───────────────────────────────────────────────────────────────────────────────────────────── ─────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────── ─────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────── No sessionR
Dracula
Amplifier TUI …/ANext/amplifier-tui How do I set up a Python virtual environment? ▶ Thinking: The user is asking about Python virtual environments... ▶ Tool: bash Virtual Environments Create one with the built-in venv module: ───────────────────────────────────────────────────────────────────────────────────────────── python3 -m venv .venv source .venv/bin/activate ───────────────────────────────────────────────────────────────────────────────────────────── Your prompt will show (.venv) when active. Install packages normally with pip install. Tip: Use uv for faster package management: ───────────────────────────────────────────────────────────────────────────────────────────── uv venv .venv uv pip install requests ───────────────────────────────────────────────────────────────────────────────────────────── What about conda? ▼ Thinking: Conda is a different ecosystem... Conda is a different ecosystem. I should compare it with venv/uv. Conda creates isolated environments with their own Python: ───────────────────────────────────────────────────────────────────────────────────────────── conda create -n myenv python=3.11 conda activate myenv▆▆ ───────────────────────────────────────────────────────────────────────────────────────────── ─────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────── ─────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────── No sessionR
Monokai
Even More Looks
Amplifier TUI …/ANext/amplifier-tui How do I set up a Python virtual environment? ▶ Thinking: The user is asking about Python virtual environments... ▶ Tool: bash Virtual Environments Create one with the built-in venv module: ───────────────────────────────────────────────────────────────────────────────────────────── python3 -m venv .venv source .venv/bin/activate ───────────────────────────────────────────────────────────────────────────────────────────── Your prompt will show (.venv) when active. Install packages normally with pip install. Tip: Use uv for faster package management: ───────────────────────────────────────────────────────────────────────────────────────────── uv venv .venv uv pip install requests ───────────────────────────────────────────────────────────────────────────────────────────── What about conda? ▼ Thinking: Conda is a different ecosystem... Conda is a different ecosystem. I should compare it with venv/uv. Conda creates isolated environments with their own Python: ───────────────────────────────────────────────────────────────────────────────────────────── conda create -n myenv python=3.11 conda activate myenv▆▆ ───────────────────────────────────────────────────────────────────────────────────────────── ─────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────── ─────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────── No sessionR
Gruvbox
Amplifier TUI How do I set up a Python virtual environment?▁▁ ▶ Thinking: The user is asking about Python virtual environments... ▶ Tool: bash Virtual Environments Create one with the built-in venv module: ────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────── python3 -m venv .venv source .venv/bin/activate ────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────── Your prompt will show (.venv) when active. Install packages normally with pip install. Tip: Use uv for faster package management: ────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────── uv venv .venv uv pip install requests ────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────── What about conda? ▼ Thinking: Conda is a different ecosystem... Conda is a different ecosystem. I should compare it with venv/uv. Conda creates isolated environments with their own Python: ────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────── conda create -n myenv python=3.11 conda activate myenv ────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────── Use conda when you need non-Python dependencies (C libraries, CUDA). Focus mode ON (F11 or /focus to exit) ─────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────── ───────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────
Focus Mode
Focus mode hides everything except what matters: the conversation.
Make the Invisible Visible
Watch your agents work
Delegating... +-- foundation:explorer (surveying auth/) [12s, 3.2k tokens] RUNNING +-- python-code-intel (tracing calls) [8s, 1.1k tokens] RUNNING +-- zen-architect (reviewing design) [pending] QUEUED

Live Delegation Trees

See which agents are running, what they're doing, and how deep the delegation goes — in real time.

Token Consumption

Track token usage per agent. Know which sub-agents are expensive and which are efficient.

No Other Tool Shows This

No other terminal AI interface visualizes agent orchestration. This is unique to Amplifier TUI.

Know Your Limits
See exactly where your tokens go
Context: [=============================----] 78% (156k / 200k) System prompt: @@@@ 12% (24k) Conversation: @@@@@@@@@@@@ 38% (76k) Tool results: @@@@@@@@ 24% (48k) Available: ---- 16% (32k)

Answer the #1 Question

Every AI tool user asks "why is it forgetting things?" This answers that question visually, in real-time. See your context window fill up, know when you're running low, and act before it's too late.

Explore Alternatives
Fork conversations like git branches
main |- msg 1: "Refactor the auth module" |- msg 2: response... | +-- branch/try-jwt "What about using JWT instead?" | |- msg 3: response... | +-- branch/try-oauth "Show me the OAuth approach" |- msg 3: response...

Fork at Any Message

Create a branch from any point in the conversation to explore a different direction.

Switch & Compare

Jump between branches instantly. Compare responses side by side to pick the best approach.

Merge Results Back

Found what you want? Merge it back to your main conversation thread.

No other AI interface lets you explore alternative conversation paths and compare them.
Compare Models
Same prompt, two models, side by side

Simultaneous Requests

Send the same prompt to Claude and GPT-4 at the same time. Watch both responses stream in.

Pick the Winner

Compare quality, speed, and style. Choose the response you prefer and continue with that model.

Perfect for Evaluation

Running model evaluations? A/B testing makes it effortless to compare outputs on real prompts.

Any Provider

Works with any provider Amplifier supports — Anthropic, OpenAI, Azure, local models, and more.

Learn & Demo
Play back any session like a movie

Adjustable Speed

0.5x for studying, 2x for review, instant for jumping to the end. Full playback control.

Timeline Scrubber

Jump to any point in the session. See messages appear with their original timing.

Great for Demos

Show stakeholders how a complex agent session unfolded. Better than screenshots.

Learn from Past Work

Review past sessions to understand what worked, what didn't, and how to prompt better.

And Everything Else
78+ commands and counting
Amplifier TUI ▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔                    Keyboard Shortcuts ────────────────────────────────────────────────────────────  NAVIGATION ───────────────────────────────────────────────   Ctrl+↑/↓         Scroll chat up/down   Ctrl+Home/End    Jump to top/bottom   Home/End          Top/bottom (input empty)   ↑/↓              Browse prompt history   Ctrl+B           Toggle sidebar   Tab              Cycle focus▁▁   F11              Focus mode  CHAT ─────────────────────────────────────────────────────   Enter            Send message   Shift+Enter      New line (also Ctrl+J)   Alt+M            Toggle multiline mode                     (ML: Enter=newline, Shift+Enter=send)   Escape           Cancel AI streaming   Ctrl+C           Cancel AI response   Ctrl+L           Clear chat   Ctrl+A           Toggle auto-scroll  SEARCH & EDIT ────────────────────────────────────────────   Ctrl+F           Find in chat (search bar)   Ctrl+R           Reverse history search (Ctrl+S fwd)   Ctrl+G           Open external editor   Ctrl+Y           Copy last AI response   Ctrl+M           Bookmark last response   Ctrl+S           Stash/restore draft  VIM BOOKMARKS (normal mode) ──────────────────────────────   Ctrl+B           Toggle bookmark on last msg                 Jump to next bookmark  SESSIONS ─────────────────────────────────────────────────   Ctrl+N           New session   Ctrl+T           New tab   F2               Rename current tab ▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁
Vim mode
Multi-tab (9 tabs)
Split view
Clipboard ring
Session tags
Inline diffs
Plugin system
Smart /include
Recipe pipeline viz
Heatmap dashboard
Origin Story
Built by AI, for AI users

Session 1: The Marathon

62 hours, 9 prompts, 147 commits. One autonomous session built the entire TUI from scratch.

Session 2: The Sprint

14 features overnight, 659 new tests. A structured feature backlog executed autonomously.

The Result

922 tests, 0 errors, 2 sessions. A production-quality terminal UI built almost entirely by AI agents.

Read the full marathon story →
Sources
Research Methodology

Data as of: February 2026

Feature status: Active

Research performed:

Gaps: Exact hour count for marathon session is approximate based on session timestamps; commit counts from git log

Primary contributors: ramparte (sole author, ~100% commits)

Try It
AI deserves a real interface.
Terminal-native. Agent-aware. Beautiful.
The TUI: github.com/ramparte/amplifier-tui
Amplifier: github.com/microsoft/amplifier
See your agents. Understand your context. Branch your conversations.