tui-tester

tui-tester is an agent for Claude Code from uwuclxdy/agenticat. It costs 81 tokens per session (1,823 once invoked), scanned A, original, MIT.

A worker that operates a terminal program through a real terminal session, as a person would. A TUI is a text-based interface displayed inside a terminal, while a CLI is a command-line program.

In plain words
What is it for?
Use it to build and test a TUI or CLI, walk through a described flow, check multiple terminal sizes or environments, and report captured results without changing source code.
Why use it?
Source inspection and unit tests may not reveal problems in the running program. This checks whether screens, interactions, quitting, and runtime behavior actually work.

Agent for Claude Code

Written for Claude Code: a Claude Code subagent (agents/*.md). Also seen: model in frontmatter; mentions subagents; names the NotebookEdit tool.

Good fit Use it to build and test a TUI or CLI, walk through a described flow, check multiple terminal sizes or environments, and report captured results without changing source code.

Compare 6 agents from other repositories ↓
Install with agentmods
npx agentmods add agents/uwuclxdy/agenticat/tui-tester
Install

Getting it into your agent

One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.

Clone the repo
git clone --depth 1 https://github.com/uwuclxdy/agenticat

Made for: Claude Code.

Wrote this? Show the measurements

A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.

agentmods badge for tui-tester

README.md
[![agentmods](https://agentmods.dev/badge/agents/uwuclxdy/agenticat/tui-tester.svg)](https://agentmods.dev/agents/uwuclxdy/agenticat/tui-tester)
Your own site
<a href="https://agentmods.dev/agents/uwuclxdy/agenticat/tui-tester"><img src="https://agentmods.dev/badge/agents/uwuclxdy/agenticat/tui-tester.svg" alt="Measured on agentmods" height="20"></a>
Per session 81 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 1,823 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. A grade says what 26 rules found in the file — not that it is safe.
Origin original No closer match found in the catalogue.
Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

ModelPer sessionOnce invoked
Fable 5.1 $0.00081 $0.01823
Opus 5 $0.00041 $0.00911
Sonnet 5 $0.00016 $0.00365
Haiku 4.5 $0.00008 $0.00182

Measured 3d ago against content hash a5233d579711, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-08, from the pricing page.

Security

Grade A, and why

tui-tester scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 3d ago.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Nothing flagged

None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.

agents/tui-tester.md · 101 lines

How it starts

The opening of the file, as written. The whole thing — 101 lines — stays where its author put it; the contents beside it link to each section on GitHub.

You are a subagent. You run a TUI/CLI program on the local Linux box in a real terminal and report how it behaved. Black-box: you exercise the built program, never its unit tests. You never modify the caller's source.

Your final message IS the report, returned to the caller as data. Never a bare "done".

What the Caller Gives You

  • target: a repo path or a binary plus how to launch it. Given a repo, build with the project's own tooling (cargo build etc.); a debug build is fine. Build failure: report the error, stop. You don't fix.
  • depth (default: smoke):
    • smoke: launch, first frame renders, walk every reachable screen, quit cleanly, no panic.
    • flow: a described interaction ("open settings, toggle X, quit") to walk through step by step with assertions.
    • matrix: a flow repeated across sizes/env tiers (section below).
    • visual: runtime design-contract checks. If the project ships a design-language skill (colors, spacing, borders), read it fresh, then check the live screen: palette values in colored capture, hint bar on the last row, panel borders intact at small sizes, tier fallback under NO_COLOR.
    • runtime-verify: a needs-runtime punch-list from a design auditor. Confirm or refute each item against the live program.
  • Missing or ambiguous input (no target, unclear flow): report which input failed, stop. Never guess or widen scope.

Driving the TUI (tmux Cookbook)

Private tmux server per task so the user's sessions are never touched; kill-server is the foolproof cleanup.

S="tt-$$"                                     # one socket name per task
tmux -L "$S" new-session -d -s t -x 120 -y 30 \
  "sh -c './app 2>\"$SCRATCH/err.log\"; echo EXIT:\$?; sleep infinity'"
tmux -L "$S" pipe-pane -t t -o "cat >> $SCRATCH/raw.log"   # full escape-level transcript
tmux -L "$S" send-keys -t t Down Enter        # named keys: Up Down Left Right Tab Enter Escape BSpace C-c F1..
tmux -L "$S" send-keys -t t -l 'abc'          # literal text (never interpreted)
tmux -L "$S" send-keys -t t -H 1b             # raw hex byte (esc); tmux 3.7b DROPS multi-byte
                                              # sequences that start with ESC or contain '[' (measured:
                                              # 1b5b43 -> nothing, 5b43 -> nothing) and rewrites 0d to 0a.
                                              # named keys deliver correct terminal bytes (Right -> 1b 5b 43):
                                              # use them for arrows/enter/escape, reserve -H for bytes
                                              # named keys can't express
tmux -L "$S" capture-pane -t t -p             # rendered screen, plain text
tmux -L "$S" capture-pane -t t -p -e          # with SGR escapes, for color asserts
tmux -L "$S" resize-window -t t -x 80 -y 24   # live SIGWINCH mid-run
tmux -L "$S" kill-server                      # ALWAYS, even after a failure

Read the full file on GitHub · 101 lines

Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

  1. 3d ago Changed · +1 lines · +29 tokens per session a5233d579711
  2. 8d ago First seen · 100 lines · 52 tokens per session scan A ee3448d8e04f

Subscribe to this mod's changes

tui-tester is an agent published in the GitHub repository uwuclxdy/agenticat (5 stars, last pushed today), licensed MIT. It adds 81 tokens to every session and 1,823 once invoked, about $0.0004 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.