fix: aid <workspace> goes straight to the prompt editor again

CHANGELOG.md

@@ -7,6 +7,8 @@ and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0
 
 ## [Unreleased]
 
+## [0.59.1] - 2026-10-01
+
 ### Fixed
 
 - **`rm` no longer counts a commit and the commit that reverts it** (#664). Neither has a
@@ -19,6 +21,10 @@ and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0
   and so does a second commit that grew from it, such as a branch or the stash. A merge,
   a root commit, an empty commit and a revert of a commit that another rule cleared
   still count, and so does an earlier draft of a commit that was edited later.
+- **`aid <workspace>` goes straight to the prompt editor again** (#666). 0.59.0 asked for the
+  agent, the model and the effort on every promptless launch on a terminal. Only a line
+  with no workspace, such as a bare `aid` or `aid --codex`, is asked now. A line that
+  names its workspace starts the agent on its defaults and the line's flags, as 0.58.0 did.
 
 ## [0.59.0] - 2026-09-30
 

README.md

@@ -19,7 +19,7 @@ one argument instead of a clone, a config file and a build command.
 [![GitHub pull-requests merged](https://badgen.net/github/merged-prs/blooop/devlaunch)](https://github.com/blooop/devlaunch/pulls?q=is%3Amerged)
 [![GitHub release](https://img.shields.io/github/release/blooop/devlaunch.svg)](https://GitHub.com/blooop/devlaunch/releases/)
 [![PyPI](https://img.shields.io/pypi/v/devlaunch)](https://pypi.org/project/devlaunch/)
-[![Conda](https://img.shields.io/badge/conda-v0.59.0-brightgreen?logo=anaconda)](https://prefix.dev/channels/blooop/packages/devlaunch)
+[![Conda](https://img.shields.io/badge/conda-v0.59.1-brightgreen?logo=anaconda)](https://prefix.dev/channels/blooop/packages/devlaunch)
 [![License](https://img.shields.io/github/license/blooop/devlaunch)](https://opensource.org/license/mit/)
 [![Platform](https://img.shields.io/badge/platform-linux--64-blue)](https://github.com/blooop/devlaunch/releases)
 [![Pixi Badge](https://img.shields.io/endpoint?url=https://raw.githubusercontent.com/prefix-dev/pixi/main/assets/badge/v0.json)](https://pixi.sh)
@@ -341,7 +341,7 @@ clone, and [docs/cleanup.md](docs/cleanup.md) says what it carries one past and
 
 ```bash
 $ dl --version
-dl 0.59.0
+dl 0.59.1
 ```
 
 `--devcontainer <variant|path>` picks a non-default `devcontainer.json`. A bare name means
@@ -411,10 +411,11 @@ With no workspace on the line, `aid` on a terminal lets you pick one of your wor
 does. With no prompt on the line, `aid` starts the container booting and asks for the prompt while it
 does. Type it free of shell quoting, with no escaping and no history expansion eating a `!`. A
 paste keeps its line breaks and can be any length, Alt-Enter or Ctrl-J adds a line, and an empty
-Enter starts the agent's plain session. Before the prompt it asks for the agent (one row per Claude
-login, then `codex` and `gemini`), the model and the effort, each in a picker that lists your
-recent choices first, so one Enter repeats the last launch. A flag on the line skips its picker,
-and Esc stops the boot. See [docs/cli.md](docs/cli.md#the-pickers-ahead-of-the-prompt). Piping
+Enter starts the agent's plain session. A bare `aid`, which picks its workspace, also asks before
+the prompt for the agent (one row per Claude login, then `codex` and `gemini`), the model and the
+effort, each in a picker that lists your recent choices first, so one Enter repeats the last
+launch. A flag on the line skips its picker, and Esc stops the boot. `aid <workspace>` asks for
+none of them and goes straight to the prompt. See [docs/cli.md](docs/cli.md#the-pickers-ahead-of-the-prompt). Piping
 stdin or setting `DEVLAUNCH_NO_TTY=1` skips the question and launches one-shot, so scripts behave
 as they always have.
 

docs/cli.md

@@ -759,8 +759,14 @@ Remote Control. A value that is missing, empty or starts with `-` is refused too
 
 ## The pickers ahead of the prompt
 
-`aid <workspace>` with no prompt on a terminal asks for up to three settings before
-it opens the prompt editor. The workspace boots in the background the whole time.
+`aid` with no workspace and no prompt on a terminal asks for up to three settings
+after the workspace picker and before it opens the prompt editor. The workspace
+boots in the background the whole time. A line of flags with no workspace, such as
+`aid --codex`, is asked the same way.
+
+`aid <workspace>` asks for none of them. A line that names its workspace said what
+it wanted, so it goes straight to the prompt editor, and every setting its flags
+left open stays at the agent's default.
 
 1. **The agent**, with one row per Claude login and one row for each other agent:
 
@@ -797,7 +803,7 @@ Where the name is part of a listed row, such as `gpt-5.5` beside `gpt-5.5-codex`
 Alt-Enter uses the text as typed instead of the row.
 
 A setting that a flag on the line already gave is not asked for, so
-`aid --model opus --effort max <ws>` goes straight to the prompt. A line with a
+`aid --model opus --effort max` asks for the workspace and the agent only. A line with a
 prompt on it asks for nothing, as before.
 
 **Esc cancels the launch.** In a picker Esc and Ctrl-C are keys, not signals, so

rust/Cargo.toml

@@ -11,7 +11,7 @@ members = [
 # The single source of the version (docs/rust-rewrite-plan.md: cutover ships
 # 0.1.0, version read from Cargo.toml).
 [workspace.package]
-version = "0.59.0"
+version = "0.59.1"
 edition = "2024"
 license = "MIT"
 repository = "https://github.com/blooop/devlaunch"

rust/aid/src/interactive.rs

@@ -162,6 +162,17 @@
     }
 }
 
+/// Whether the agent, model and effort pickers run before the editor.
+///
+/// Only a bare `aid`, or a line of flags with no workspace, asks. A line that
+/// names its workspace already said what it wanted, so it gets the editor alone
+/// and every setting it left open stays the agent's default.
+#[derive(Clone, Copy)]
+pub(crate) enum Pickers {
+    Ask,
+    Skip,
+}
+
 /// How the interactive flow ended.
 pub(crate) enum Collected {
     /// Launch this line, once the boot beside it, if any, has finished.
@@ -187,10 +198,12 @@
 ///
 /// `argv` and `environment` are what `parsed` was parsed from: the agent picker
 /// builds its rows by parsing the line again with each row's flags in front.
+/// `pickers` says whether those pickers run at all.
 pub(crate) fn collect_prompt(
     parsed: AidArgs,
     argv: &[String],
     environment: Environment<'_>,
+    pickers: Pickers,
 ) -> Collected {
     let promptless_agent = matches!(
         &parsed.task,
@@ -216,9 +229,15 @@
     // The pickers come after the boot has started, so the minute they take is
     // also spent booting. Nothing they choose reaches the boot: `up` takes no
     // model, and `--claude-profile` is read when the session starts, not at `up`.
-    let Some(parsed) = settle(parsed, argv, environment) else {
-        boot.cancel();
-        return Collected::Cancelled;
+    let parsed = match pickers {
+        Pickers::Skip => parsed,
+        Pickers::Ask => {
+            let Some(parsed) = settle(parsed, argv, environment) else {
+                boot.cancel();
+                return Collected::Cancelled;
+            };
+            parsed
+        }
     };
     banner(&parsed);
     match dl::read_prompt() {

rust/aid/src/main.rs

@@ -150,14 +150,22 @@
     // id is added to the end of the line, after the leading flags, which is where a
     // typed spec goes, and from here on the line is that longer one, because the
     // agent picker parses it again.
+    //
+    // Only that last pick, a workspace for a line that named none, earns the agent,
+    // model and effort pickers: a line that names its workspace goes straight to
+    // the editor.
     let with_pick: Vec<String>;
-    let (argv, parsed) = match rewrite::parse_aid_args(argv, environment) {
-        Ok(rewrite::Line::Ready(parsed)) => (argv, parsed),
+    let (argv, parsed, pickers) = match rewrite::parse_aid_args(argv, environment) {
+        Ok(rewrite::Line::Ready(parsed)) => (argv, parsed, interactive::Pickers::Skip),
         // `aid resume` with no workspace. The pick comes before everything below,
         // which is all about one named workspace, so from here on this line is an
         // `aid resume <id>` like any other.
         Ok(rewrite::Line::Unpicked(unpicked)) => match dl::pick_workspace() {
-            Ok(workspace_id) => (argv, unpicked.picked(workspace_id)),
+            Ok(workspace_id) => (
+                argv,
+                unpicked.picked(workspace_id),
+                interactive::Pickers::Skip,
+            ),
             Err(code) => return code,
         },
         Err(UsageError::NoWorkspace)
@@ -169,7 +177,9 @@
             };
             with_pick = argv.iter().cloned().chain([spec]).collect();
             match rewrite::parse_aid_args(&with_pick, environment) {
-                Ok(rewrite::Line::Ready(parsed)) => (with_pick.as_slice(), parsed),
+                Ok(rewrite::Line::Ready(parsed)) => {
+                    (with_pick.as_slice(), parsed, interactive::Pickers::Ask)
+                }
                 // A line that names its workspace is never unpicked.
                 Ok(rewrite::Line::Unpicked(_)) => {
                     eprintln!("{}", refusal(&UsageError::NoWorkspace));
@@ -199,7 +209,7 @@
         Ok(spec) => parsed.with_spec(spec),
         Err(code) => return code,
     };
-    let (parsed, boot) = match interactive::collect_prompt(parsed, argv, environment) {
+    let (parsed, boot) = match interactive::collect_prompt(parsed, argv, environment, pickers) {
         interactive::Collected::Launch(parsed, boot) => (*parsed, boot),
         // 130, the code a Ctrl-C at the prompt editor ends with, because a cancel
         // in a picker is the same request made with a different key.
@@ -408,14 +418,15 @@
 reopens its latest session.
 
 With no workspace on a terminal, aid lets you pick one of your workspaces, as
-dl does.
+dl does. Then it asks for each setting the line left open: the agent (one row
+per Claude login, then each other agent), the model and the effort. Each
+picker lists your recent choices first, so one Enter repeats the last launch.
+Type a name that is not listed to use it. Alt-Enter uses the text as typed
+where it is part of a listed name. A line that names its workspace asks for
+none of these, and the agent starts on its defaults and the line's flags.
 
 With no prompt on a terminal, aid boots the workspace in the background and
-asks for the prompt while it does. First it asks for each setting the line
-left open: the agent (one row per Claude login, then each other agent), the
-model and the effort. Each picker lists your recent choices first, so one
-Enter repeats the last launch. Type a name that is not listed to use it.
-Alt-Enter uses the text as typed where it is part of a listed name.
+asks for the prompt while it does.
 
 Then type the prompt free of shell quoting and press Enter to launch. A paste
 keeps its line breaks, and Alt-Enter or Ctrl-J adds a line. An empty Enter