auto-yes 0.9.1

Automatically respond 'yes' to interactive CLI prompts via PTY proxy

auto-yes

Automatically respond yes to interactive CLI prompts.

auto-yes wraps your shell or a single command inside a PTY proxy, watches the output for common interactive prompts ([y/n], Continue?, etc.) and injects the appropriate response — all without stealing your stdin or breaking colours, progress bars or tab-completion.

Installation

pip install auto-yes

On Windows the tool falls back to a basic pipe-based mode. For full PTY support install the optional extra:

pip install auto-yes[windows]

Quick start

Wrap an AI CLI tool (recommended)

The simplest way: auto-yes <profile> [args...] wraps a tool in a single process with auto-responses enabled and verbose output on by default:

auto-yes claude "fix the tests"        # runs: claude "fix the tests"
auto-yes cursor chat "fix the bug"     # runs: agent chat "fix the bug"
auto-yes aider --model gpt-4          # runs: aider --model gpt-4
auto-yes copilot                       # runs: gh copilot
auto-yes amazonq chat "help me"        # runs: q chat "help me"

The profile name is mapped to the real binary automatically (e.g. cursor runs agent, copilot runs gh copilot). Run auto-yes list to see all available profiles and their real commands.

You can also use the short alias ay (same behavior as auto-yes):

ay cursor chat "fix the bug"
ay --on --cli claude

Shell session mode

Start a persistent shell where every prompt is answered automatically:

auto-yes --on --cli claude
# you are now inside an auto-yes shell
exit                       # leave the session

Single command mode

Wrap one command without entering a persistent session:

auto-yes run -- apt install nginx
auto-yes run --cli codex -- codex "fix tests"

Check status

auto-yes status        # → "active" or "inactive"

How it works

User terminal  ←→  PTY proxy (auto-yes)  ←→  Child process
                    │
                    ├─ reads child stdout
                    ├─ strips ANSI escapes
                    ├─ matches prompt patterns
                    ├─ injects "y\n" / "yes\n" / "\n"
                    └─ forwards everything to user

1. A pseudo-terminal pair (master/slave) is created. The child process (your shell or command) is spawned on the slave side, so it believes it is running in a real terminal.

2. The parent process sits on the master side, using select() to multiplex user input and child output in real time.

3. Each chunk of child output is appended to a rolling buffer. The buffer is cleaned (ANSI stripped, carriage-returns resolved) and checked against a set of regex patterns.

4. When a match is found on the last visible line (i.e. the process is actually waiting for input), the configured response is written into the master fd — the child sees it as if the user typed it.

5. A cooldown timer prevents responding twice to the same prompt.

6. SIGWINCH is forwarded so terminal resizes propagate correctly.

CLI reference

auto-yes <profile> [ARGS...]       wrap an AI CLI tool directly (recommended)
auto-yes --on [OPTIONS]            start an auto-yes shell session
auto-yes run [OPTIONS] -- CMD...   run a single command with auto-yes
auto-yes list, -l, --list          list all available CLI profiles
auto-yes patterns [CATEGORY...]    list prompt patterns (optionally filtered)
auto-yes add-pattern PATTERN       persist a custom regex pattern
auto-yes del-pattern PATTERN       remove a custom pattern
auto-yes status                    check if auto-yes is active
auto-yes --off                     exit info

Options (for --on and run)

Flag	Description	Default
--response TEXT	Text to send when a prompt is detected	y
--cooldown FLOAT	Seconds between auto-responses	0.5
--verbose, -v	Print a notice each time auto-yes responds	off
--pattern REGEX	Extra prompt pattern (repeatable)	—
--cli NAME	AI CLI profile to load (repeatable, or all)	—

Pattern categories

Patterns are organized by category for maintainability. Each AI CLI profile only loads its own patterns — no broad generic rules that might auto-confirm dangerous prompts like pip uninstall. The generic category (universal [y/n], Continue?, etc.) is opt-in via --cli generic if you want it.

generic (opt-in via --cli generic)

Type	Examples
Bracket choices	[y/n], [Y/n], (y/N), [yes/no], ([y]/n), (y)
Styled choices	[Y]es / [N]o, bare y/n:
Question sentences	Continue?, Proceed?, Are you sure?, Would you like to…?
Safe action prompts	Allow?, Approve?, Accept?, Download?, Enable?, Create?
Excluded (dangerous)	Overwrite?, Delete?, Remove?, Upgrade?, Update?, Merge?, Restart?, Reboot?, Install? (use --pattern to add manually)
Agreement prompts	Do you agree?, Do you accept?, Accept the license, Agree to the terms
Need/require prompts	Need to install?, Required?, Necessary?
Package managers	Is this ok [y/d/N] (pip excluded to avoid accidental uninstall)
SSH fingerprint	continue connecting (yes/no/[fingerprint])? → responds yes
Terraform	Only 'yes' will be accepted → responds yes
Full-word yes	Type 'yes' to continue, Type 'YES' to confirm, Enter 'yes' to proceed
Press enter/key	Press Enter to continue, Press any key to continue → responds with empty
Default value	(default: Y), [default=yes] → responds with empty (accept default)

AI CLI profiles

Profile	Command	Tool	Key patterns
claude	claude	Anthropic Claude Code	1. Yes, I trust this folder, Yes, allow, API key prompt
gemini	gemini	Google Gemini CLI	1. Allow once, 2. Allow for this session, 3. Always allow
codex	codex	OpenAI Codex CLI	1. Approve and run now, 1. Yes, allow Codex to work
copilot	gh copilot	GitHub Copilot CLI	1. Yes, proceed, Allow Copilot to run
cursor	agent	Cursor Agent CLI	→ Run (once) (y), → Run (always) (a), Trust this workspace
grok	grok	xAI Grok CLI	1. Yes
auggie	auggie	Augment Code CLI	[Y] Enable indexing
amp	amp	Sourcegraph Amp CLI	Approve
aider	aider	Aider AI Coding	(Y)es/(N)o, Run shell command?, Add … to the chat?
openhands	openhands	OpenHands AI Agent	Do you want to execute this action?, Approve
windsurf	windsurf	Codeium Windsurf	Accept changes?, Run this command?
qwen	qwen	Alibaba Qwen Code	1. Yes, Approve execution?
amazonq	q	Amazon Q Developer	Do you approve this action?, Accept suggestion?

# recommended: wrap directly (only loads tool-specific patterns)
auto-yes claude "fix the tests"
auto-yes cursor chat "fix the bug"

# opt-in generic patterns if needed
auto-yes --on --cli claude --cli generic

# load all profiles including generic
auto-yes --on --cli all

# inspect patterns for a specific profile
auto-yes patterns claude codex

# list all profiles and their real commands
auto-yes list

Custom patterns

# persist a pattern across sessions
auto-yes add-pattern 'accept license\?'

# one-off extra pattern
auto-yes run --pattern 'custom_prompt\?' -- ./my-script.sh

Adding a new AI CLI profile

Add a new entry to REGISTRY in src/auto_yes/patterns.py:

_MY_TOOL = {
    "description": "My AI Tool CLI",
    "command": ["my-tool"],              # real binary name (list of strings)
    "patterns": [
        (r"pattern_regex_here", None),    # respond with default
        (r"another_pattern", "yes"),      # respond with "yes"
    ],
}

REGISTRY["my-tool"] = _MY_TOOL

The command field maps the profile name to the real binary. For example, the cursor profile has "command": ["agent"], and copilot has "command": ["gh", "copilot"]. No other file needs to change.

Configuration

Stored at ~/.config/auto-yes/config.json (Linux/macOS) or %APPDATA%\auto-yes\config.json (Windows):

{
  "custom_patterns": [],
  "response": "y",
  "cooldown": 0.5,
  "verbose": false
}

Platform support

Platform	Method	Notes
Linux	pty + select	Full PTY, zero external deps
macOS	pty + select	Full PTY, zero external deps
Windows	pywinpty	Install auto-yes[windows]
Windows (fallback)	subprocess pipes	Works but no true PTY

Comparison with yes(1)

	yes | cmd	auto-yes run -- cmd
Prompt detection	None (floods stdin)	Smart regex matching
User interaction	Impossible	Preserved
PTY	No (pipe)	Yes
Colours / progress bars	Often broken	Preserved
Cross-platform	Unix only	Unix + Windows

Python API

from auto_yes.runner import Runner
from auto_yes.patterns import get_command

# generic patterns only
runner = Runner(response="y", cooldown=0.5, verbose=True)
exit_code = runner.run_command(["apt", "install", "nginx"])

# with AI CLI profile (using real binary name from registry)
runner = Runner(categories=["generic", "cursor"], verbose=True)
cmd = get_command("cursor")          # -> ["agent"]
exit_code = runner.run_command(cmd + ["chat", "fix the bug"])

Development

pip install -e ".[dev]"
pytest

License

MIT