What is agent-tui AGENTS.md?

A project guide for agent-tui, a command-line tool that lets AI agents control terminal applications through managed terminal sessions. It describes the code layers, repository structure, and rules such as structured errors and dependency direction.

Use it when navigating the agent-tui repository, adding features, fixing bugs, handling errors, or checking whether code follows its architecture and async rules.

It gives agents the boundaries and conventions needed to change the Rust project without placing code in the wrong layer or violating its safety rules.

agent-tui AGENTS.md — Instructions file by pproenca

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.

agentmods

npx agentmods add instructions/pproenca/agent-tui/agents-md

Clone the repo

git clone --depth 1 https://github.com/pproenca/agent-tui

Made for: Codex, OpenCode.

Per session 1,311 This file is loaded in full into every session.

When invoked 1,311 The same file — it is already loaded in full.

Security scan A 0 findings. Scan, not verified.

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.

Model	Per session	Once invoked
Fable 5	$0.01311	$0.01311
Opus 5	$0.00656	$0.00656
Sonnet 5	$0.00262	$0.00262
Haiku 4.5	$0.00131	$0.00131

Measured 2d ago against content hash f7ae6c3f31f3, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

agent-tui AGENTS.md 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 2d 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.md · 100 lines

How it starts

The opening of the file, as written. The whole thing — 100 lines — stays where its author put it; the contents beside it link to each section on GitHub.

agent-tui — Agent Guide

CLI tool that enables AI agents to programmatically drive terminal applications. Manages PTY sessions via a background daemon with JSON-RPC WebSocket API and embedded web UI.

Non-Negotiable Rules

Dependencies flow forward only. common → domain → usecases → adapters/infra → app → facade. No reverse imports. Validated by just boundaries and CI.
No panics in production code. unwrap_used and expect_used are denied workspace-wide. Use ?, anyhow::Context, or thiserror types. Tests may unwrap.
No blocking in async context. std::thread::sleep, unbounded channels, and std::process::exit (except main.rs) are banned via Clippy disallowed methods.
All errors use structured types. Domain errors use thiserror; infra/app errors use anyhow with .context(). Never swallow errors silently.
Facade crate is a shell. cli/crates/agent-tui/ contains only main.rs, lib.rs, and bin/*.rs. All logic lives in layer crates.

Repository Map

Path	What it contains
`ARCHITECTURE.md`	Domain boundaries, dependency graph, layer descriptions, where new code goes
`docs/design-docs/`	Design history, core beliefs, architectural decisions
`docs/api/`	OpenAPI + AsyncAPI specs, example clients (Rust, JS)
`docs/ops/`	Process model, deployment patterns
`docs/agents/`	Matt Pocock engineering workflow config: issue tracker, labels, project board, and domain docs
`docs/cli/`	Auto-generated CLI reference (do not hand-edit)
`cli/docs/architecture/`	Clean Architecture target state and dependency matrix
`cli/docs/plans/`	Historical refactor/restructure plans
`.harness/`	Machine-readable harness config (domains, principles, enforcement, quality)
`skills/`	Agent skill definitions and references

Tech Stack

Rust 1.88+ (2024 edition) — Tokio async runtime, Axum WebSocket server, Clap CLI
TypeScript/Bun — Web UI with xterm.js terminal emulator
Build: Cargo workspace (8 crates) + just task runner
Key deps: portable-pty, tracing, serde, crossterm

Read the full file on GitHub · 100 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.

2d ago First seen · 100 lines · 1,311 tokens per session scan A f7ae6c3f31f3

Subscribe to this mod's changes

agent-tui AGENTS.md is an instructions file published in the GitHub repository pproenca/agent-tui (113 stars, last pushed 3d ago), licensed MIT. It adds 1,311 tokens to every session, about $0.0066 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 2026-08-30.

Debugging Languages Planning ai-agents cli json-rpc pty rust terminal-automation

agent-tui AGENTS.md

Getting it into your agent

What it costs to keep this loaded

Grade A, and why

Nothing flagged

How it starts

agent-tui — Agent Guide

Non-Negotiable Rules

Repository Map

Tech Stack

What this file has done since we first saw it

Other instructions, from other repositories

spacevibe-deck AGENTS.md

Codex-web AGENTS.md

tower-mcp AGENTS.md

zashiki CLAUDE.md

Codex-web CLAUDE.md

spacevibe-deck CLAUDE.md