feat(coding-agent): added toggle for launch tool

- Added a configuration setting `launch.enabled` to control the availability of the project-scoped launch tool.
- Integrated the setting into the bash tool and tool registry to conditionally enable or disable the launch functionality.
- Updated documentation and settings schema to include the new toggle.

docs/settings.md

@@ -451,7 +451,7 @@ tools:
 | `tools.artifactTailBytes` | number | `20` | KB of tail kept inline on spill. |
 | `tools.artifactTailLines` | number | `500` | Max tail lines kept inline on spill. |
 
-Individual built-in tools are toggled by their own keys, e.g. `bash.enabled`, `eval.py`, `eval.js`, `glob.enabled`, `grep.enabled`, `fetch.enabled`, `browser.enabled`, `astEdit.enabled`, `astGrep.enabled`, `web_search.enabled`, `inspect_image.enabled`.
+Individual built-in tools are toggled by their own keys, e.g. `bash.enabled`, `launch.enabled`, `eval.py`, `eval.js`, `glob.enabled`, `grep.enabled`, `fetch.enabled`, `browser.enabled`, `astEdit.enabled`, `astGrep.enabled`, `web_search.enabled`, `inspect_image.enabled`.
 
 ### Shell, eval, and LSP
 
@@ -481,6 +481,7 @@ lsp:
 | Key | Type | Default | Notes |
 |---|---|---|---|
 | `bash.enabled` | boolean | `true` | Enable the bash tool. |
+| `launch.enabled` | boolean | `true` | Enable the launch tool for shared long-running project processes. |
 | `bash.autoBackground.enabled` | boolean | `false` | Auto-background long-running commands. |
 | `bash.autoBackground.thresholdMs` | number | `60000` | Threshold before auto-backgrounding. |
 | `eval.py` | boolean | `true` | Python eval backend. `PI_PY=0` disables for the process. |

packages/coding-agent/CHANGELOG.md

@@ -8,7 +8,7 @@
 - Added a compact session-only model picker (Alt+P) for quick model switching without changing roles
 - Added `@` search to the Alt+P / `/switch` picker: it lists configured Ctrl+P quick roles in matching segment colors and applies the selected role's model and thinking for the current session.
 - Redesigned Agent Hub entries as two-line cards: identity (status glyph, name, agent type, parent when nested) on the left, active model + reasoning level and age right-aligned, with the task description on its own line; dropped the redundant `sub · of Main` noise
-- Added a project-scoped `launch` tool for shared long-running services and debuggers, with readiness probes, bounded logs, PTY input, restart policies, and automatic teardown after the last omp instance exits.
+- Added a project-scoped `launch` tool for shared long-running services and debuggers, with readiness probes, bounded logs, PTY input, restart policies, and automatic teardown after the last omp instance exits. Gated behind the `launch.enabled` setting (default on); when disabled the tool is withdrawn and the bash prompt drops its "use launch" guidance.
 - Added `detached` `launch` starts for standalone services that survive every omp instance and broker shutdown, then reconnect to the next broker for logs and explicit stop.
 
 ### Changed

packages/coding-agent/src/config/settings-schema.ts

@@ -3549,6 +3549,17 @@ export const SETTINGS_SCHEMA = {
 		},
 	},
 
+	"launch.enabled": {
+		type: "boolean",
+		default: true,
+		ui: {
+			tab: "tools",
+			group: "Available Tools",
+			label: "Launch",
+			description: "Enable the launch tool for supervising shared long-running project processes",
+		},
+	},
+
 	"speechgen.enabled": {
 		type: "boolean",
 		default: false,

packages/coding-agent/src/tools/bash.ts

@@ -397,7 +397,7 @@ export class BashTool implements AgentTool<typeof bashSchemaBase | typeof bashSc
 			hasGrep: isToolActive("grep", this.session.settings.get("grep.enabled")),
 			hasGlob: isToolActive("glob", this.session.settings.get("glob.enabled")),
 			hasRead: isToolActive("read", true),
-			hasLaunch: isToolActive("launch", true),
+			hasLaunch: isToolActive("launch", this.session.settings.get("launch.enabled")),
 			hasEval: isToolActive(
 				"eval",
 				evalBackends.python || evalBackends.js || evalBackends.ruby || evalBackends.julia,

packages/coding-agent/src/tools/index.ts

@@ -609,6 +609,7 @@ export async function createTools(session: ToolSession, toolNames?: string[]): P
 		if (name === "goal") return goalEnabled && goalModeActive;
 		if (name === "lsp") return enableLsp && session.settings.get("lsp.enabled");
 		if (name === "bash") return session.settings.get("bash.enabled");
+		if (name === "launch") return session.settings.get("launch.enabled");
 		if (name === "eval") return allowEval;
 		if (name === "debug") return session.settings.get("debug.enabled");
 		if (name === "todo") return !includeYield && session.settings.get("todo.enabled");

packages/coding-agent/test/tools/index.test.ts

@@ -237,6 +237,7 @@ describe("createTools", () => {
 				"astGrep.enabled": false,
 				"astEdit.enabled": false,
 				"bash.enabled": false,
+				"launch.enabled": false,
 				"web_search.enabled": false,
 				"browser.enabled": false,
 				"inspect_image.enabled": false,
@@ -246,6 +247,7 @@ describe("createTools", () => {
 		const names = tools.map(t => t.name);
 
 		expect(names).not.toContain("bash");
+		expect(names).not.toContain("launch");
 		expect(names).not.toContain("glob");
 		expect(names).not.toContain("grep");
 		expect(names).not.toContain("ast_grep");