feat(cli): Codex CLI launcher + setup commands by diegosouzapw · Pull Request #4270 · diegosouzapw/OmniRoute

diegosouzapw · 2026-06-19T11:59:20Z

Consolidates work-in-progress that was sitting uncommitted on the shared checkout into a tracked PR (per operator request during the v3.8.29 release wrap-up). Reconstructed cleanly on release/v3.8.30.

What

omniroute launch-codex — boots OmniRoute (if needed) and launches the OpenAI Codex CLI pointed at it (local or remote VPS), no manual env/config editing.
omniroute setup-codex — generates ~/.codex profile files from OmniRoute's live model catalog.
Registered in the command registry; en.json + pt-BR.json locale sections added (keeps the cli-i18n-catalog top-level parity gate green); CODEX-CLI-CONFIGURATION.md refreshed.
Separate commit: .vscode/settings.json — disables git auto-repo-detection for the nested worktrees (perf; editor-only).

Validation

both Codex command modules load + export their register fns; registry registerCommands intact
cli-i18n-catalog test 7/7 (en keys complete + pt-BR parity)
lint: 0 errors (bin/cli is eslint-ignored)

Note: drop the chore(vscode) commit in review if the editor settings aren't wanted repo-wide.

Two new CLI subcommands mirroring the `launch`/`configure` pattern, for driving the OpenAI Codex CLI against OmniRoute: - `omniroute launch-codex` — boots OmniRoute (if needed) and launches Codex CLI pointed at it (local or remote VPS), with no manual env/config editing. - `omniroute setup-codex` — generates ~/.codex profile files from OmniRoute's live model catalog. Registered in the command registry; en.json + pt-BR.json locale sections added (keeps the cli-i18n-catalog top-level parity gate green). CODEX-CLI-CONFIGURATION.md refreshed. Consolidated from work-in-progress that was uncommitted on the shared checkout; reconstructed on release/v3.8.30.

Disable VS Code's git auto-repository-detection / submodule scan / autofetch so the Source Control view stops indexing the ~44 nested repos (worktrees + _references/* + _mono_repo/*), which caused constant "validating" churn. Only the root repo is tracked. Editor-only settings; no runtime impact.

chatgpt-codex-connector · 2026-06-19T11:59:26Z

You have reached your Codex usage limits for code reviews. You can see your limits in the Codex usage dashboard.

gemini-code-assist

Code Review

This pull request introduces two new CLI commands, launch-codex and setup-codex, along with localized descriptions, documentation updates, and VS Code settings to optimize Git performance. The feedback highlights several areas for improvement: making URL parsing more robust against trailing slashes in both commands, aligning the model categorization logic in setup-codex.mjs with the updated documentation, and adding safer error handling and response parsing to prevent potential runtime crashes. Additionally, the reviewer notes that corresponding tests should be added for these new production commands to comply with the repository's style guide.

Important

The consumer version of Gemini Code Assist on GitHub is being sunset. Starting June 18, 2026, new organization installations will be blocked, and all code review activity will officially cease on July 17, 2026.
For more details on the timeline and next steps, please review the Help Documentation.

gemini-code-assist · 2026-06-19T12:01:09Z

+ */
+async function healthCheck(baseUrl, timeoutMs = 3000) {
+  try {
+    const res = await fetch(`${baseUrl.replace(/\/v1$/, "")}/api/monitoring/health`, {


If baseUrl has a trailing slash (e.g., http://localhost:20128/v1/), the regex \/\/v1$/ will not match, leaving the trailing slash and resulting in a malformed URL like http://localhost:20128/v1//api/monitoring/health.

We can make this more robust by handling optional trailing slashes.

Suggested change

const res = await fetch(`${baseUrl.replace(/\/v1$/, "")}/api/monitoring/health`, {

const cleanBase = baseUrl.replace(/\/v1\/?$/, "").replace(/\/$/, "");

const res = await fetch(`${cleanBase}/api/monitoring/health`, {

gemini-code-assist · 2026-06-19T12:01:10Z

+ */
+export async function runSetupCodexCommand(opts = {}) {
+  const port = Number(opts.port ?? process.env.PORT ?? 20128) || 20128;
+  const baseUrl = (opts.remote ?? `http://localhost:${port}`).replace(/\/v1$/, "");


Similar to the health check, if opts.remote contains a trailing slash (e.g., http://100.67.86.91:20128/v1/), the regex \/\/v1$/ will fail to strip the /v1/ path correctly. Let's make this robust against trailing slashes.

Suggested change

const baseUrl = (opts.remote ?? `http://localhost:${port}`).replace(/\/v1$/, "");

const baseUrl = (opts.remote ?? `http://localhost:${port}`).replace(/\/v1\/?$/, "").replace(/\/$/, "");

gemini-code-assist · 2026-06-19T12:01:10Z

+  const thinkingPatterns = [
+    { re: /kmc\/kimi-k2\.7/, name: "kimi-k27", ctx: 131072, compact: 112000, toolLimit: 32768 },
+    { re: /kmc\/kimi-k2\.6/, name: "kimi-k26", ctx: 131072, compact: 112000, toolLimit: 32768 },
+    { re: /glm\/glm-5\.2-max/, name: "glm52max", ctx: 131072, compact: 112000, toolLimit: 32768 },
+    { re: /glm\/glm-5\.2$/, name: "glm52", ctx: 131072, compact: 112000, toolLimit: 32768 },
+    { re: /opencode-go\/mimo-v2\.5-pro/, name: "mimo-pro", ctx: 131072, compact: 112000, toolLimit: 32768 },
+    { re: /opencode-go\/qwen3\.7-plus/, name: "qwen37plus", ctx: 32768, compact: 28000, toolLimit: 16384 },
+  ];
+
+  // ── Good models (high effort) ─────────────────────────────────────────────
+  const goodPatterns = [
+    { re: /ollamacloud\/deepseek-v4-pro/, name: "deepseek-pro", ctx: 131072, compact: 112000, toolLimit: 32768 },
+    { re: /opencode-go\/mimo-v2\.5$/, name: "mimo", ctx: 131072, compact: 112000, toolLimit: 32768 },
+  ];
+
+  // ── Simple models (no effort) ─────────────────────────────────────────────
+  const simplePatterns = [
+    { re: /ollamacloud\/gemma4:31b/, name: "gemma4", ctx: 32768, compact: 28000, toolLimit: 16384 },
+    { re: /ollamacloud\/nemotron-3-super/, name: "nemotron", ctx: 32768, compact: 28000, toolLimit: 16384 },
+    { re: /ollamacloud\/gpt-oss:20b/, name: "gptoss", ctx: 32768, compact: 28000, toolLimit: 16384 },
+  ];
+
+  // ── Fast models (low effort) ──────────────────────────────────────────────
+  const fastPatterns = [
+    { re: /ollamacloud\/deepseek-v4-flash/, name: "deepseek-flash", ctx: 65536, compact: 56000, toolLimit: 16384 },
+    { re: /ollamacloud\/gemini-3-flash/, name: "gemini-flash", ctx: 1000000, compact: 850000, toolLimit: 32768 },
+    { re: /glm\/glm-5-turbo/, name: "glm5turbo", ctx: 131072, compact: 112000, toolLimit: 16384 },
+    { re: /glm\/glm-4\.7-flash/, name: "glm47flash", ctx: 131072, compact: 112000, toolLimit: 16384 },
+  ];


There are a few discrepancies between the model categorization patterns defined here and the documentation in CODEX-CLI-CONFIGURATION.md:

kimi-k2.6: Listed under thinkingPatterns in the code (which assigns effort: "xhigh" and summary: true), but documented under ### Good models (bons) — high effort in the markdown. It should be moved to goodPatterns.

ds/deepseek-v4-pro: Documented under ### Good models (bons) — high effort with a 1M context window, but completely missing from the code patterns.

mistral/mistral-large-latest: Documented under ### Fast models (rápidos) — low effort with a 256k context window, but completely missing from the code patterns.

Let's update the patterns to align with the documentation.

const thinkingPatterns = [ { re: /kmc\/kimi-k2\.7/, name: "kimi-k27", ctx: 131072, compact: 112000, toolLimit: 32768 }, { re: /glm\/glm-5\.2-max/, name: "glm52max", ctx: 131072, compact: 112000, toolLimit: 32768 }, { re: /glm\/glm-5\.2$/, name: "glm52", ctx: 131072, compact: 112000, toolLimit: 32768 }, { re: /opencode-go\/mimo-v2\.5-pro/, name: "mimo-pro", ctx: 131072, compact: 112000, toolLimit: 32768 }, { re: /opencode-go\/qwen3\.7-plus/, name: "qwen37plus", ctx: 32768, compact: 28000, toolLimit: 16384 }, ]; // ── Good models (high effort) ───────────────────────────────────────────── const goodPatterns = [ { re: /kmc\/kimi-k2\.6/, name: "kimi-k26", ctx: 131072, compact: 112000, toolLimit: 32768 }, { re: /ollamacloud\/deepseek-v4-pro/, name: "deepseek-pro", ctx: 131072, compact: 112000, toolLimit: 32768 }, { re: /ds\/deepseek-v4-pro/, name: "deepseek", ctx: 1000000, compact: 900000, toolLimit: 65536 }, { re: /opencode-go\/mimo-v2\.5/, name: "mimo", ctx: 131072, compact: 112000, toolLimit: 32768 }, ]; // ── Simple models (no effort) ───────────────────────────────────────────── const simplePatterns = [ { re: /ollamacloud\/gemma4:31b/, name: "gemma4", ctx: 32768, compact: 28000, toolLimit: 16384 }, { re: /ollamacloud\/nemotron-3-super/, name: "nemotron", ctx: 32768, compact: 28000, toolLimit: 16384 }, { re: /ollamacloud\/gpt-oss:20b/, name: "gptoss", ctx: 32768, compact: 28000, toolLimit: 16384 }, ]; // ── Fast models (low effort) ────────────────────────────────────────────── const fastPatterns = [ { re: /ollamacloud\/deepseek-v4-flash/, name: "deepseek-flash", ctx: 65536, compact: 56000, toolLimit: 16384 }, { re: /ollamacloud\/gemini-3-flash/, name: "gemini-flash", ctx: 1000000, compact: 850000, toolLimit: 32768 }, { re: /glm\/glm-5-turbo/, name: "glm5turbo", ctx: 131072, compact: 112000, toolLimit: 16384 }, { re: /glm\/glm-4\.7-flash/, name: "glm47flash", ctx: 131072, compact: 112000, toolLimit: 16384 }, { re: /mistral\/mistral-large-latest/, name: "mistral", ctx: 262144, compact: 220000, toolLimit: 16384 }, ];

gemini-code-assist · 2026-06-19T12:01:10Z

+    });
+    if (!res.ok) throw new Error(`HTTP ${res.status} ${res.statusText}`);
+    const body = await res.json();
+    models = body.data ?? body.models ?? [];


To make this more robust against different server response formats (e.g., if the server returns a direct JSON array instead of an object, or if body is nullish), we should perform a safer check.

Suggested change

models = body.data ?? body.models ?? [];

models = Array.isArray(body) ? body : (body?.data ?? body?.models ?? []);

gemini-code-assist · 2026-06-19T12:01:10Z

+    const body = await res.json();
+    models = body.data ?? body.models ?? [];
+  } catch (err) {
+    printError(`Failed to fetch models: ${err.message}`);


If err is not an object or is null/undefined, accessing err.message will throw a TypeError and crash the process. Let's safely access the message or fall back to the string representation of the error.

Suggested change

printError(`Failed to fetch models: ${err.message}`);

printError(`Failed to fetch models: ${err?.message || String(err)}`);

gemini-code-assist · 2026-06-19T12:01:10Z

+import { registerLaunchCodex } from "./launch-codex.mjs";
+import { registerSetupCodex } from "./setup-codex.mjs";


According to the Repository Style Guide (Rule 9 under 2. Hard Rules):

Always include tests when changing production code (src/, open-sse/, electron/, bin/).

Since this PR introduces new production CLI commands in bin/cli/commands/, please ensure that corresponding unit or integration tests are added within the tests/ directory to maintain coverage.

References

Always include tests when changing production code (src/, open-sse/, electron/, bin/). ^(link)

* feat(cli): Codex CLI launcher + setup commands Two new CLI subcommands mirroring the `launch`/`configure` pattern, for driving the OpenAI Codex CLI against OmniRoute: - `omniroute launch-codex` — boots OmniRoute (if needed) and launches Codex CLI pointed at it (local or remote VPS), with no manual env/config editing. - `omniroute setup-codex` — generates ~/.codex profile files from OmniRoute's live model catalog. Registered in the command registry; en.json + pt-BR.json locale sections added (keeps the cli-i18n-catalog top-level parity gate green). CODEX-CLI-CONFIGURATION.md refreshed. Consolidated from work-in-progress that was uncommitted on the shared checkout; reconstructed on release/v3.8.30. * chore(vscode): reduce git repo-detection overhead for nested worktrees Disable VS Code's git auto-repository-detection / submodule scan / autofetch so the Source Control view stops indexing the ~44 nested repos (worktrees + _references/* + _mono_repo/*), which caused constant "validating" churn. Only the root repo is tracked. Editor-only settings; no runtime impact.

diegosouzapw added 2 commits June 19, 2026 08:59

gemini-code-assist Bot reviewed Jun 19, 2026

View reviewed changes

diegosouzapw merged commit c34d37a into release/v3.8.30 Jun 19, 2026
4 checks passed

diegosouzapw deleted the feat/codex-cli-commands branch June 19, 2026 12:36

Navigation Menu

Uh oh!

feat(cli): Codex CLI launcher + setup commands - #4270

feat(cli): Codex CLI launcher + setup commands#4270
diegosouzapw merged 2 commits into
release/v3.8.30from
feat/codex-cli-commands

diegosouzapw commented Jun 19, 2026

Uh oh!

chatgpt-codex-connector Bot commented Jun 19, 2026

Uh oh!

gemini-code-assist Bot left a comment

Uh oh!

gemini-code-assist Bot Jun 19, 2026

Uh oh!

gemini-code-assist Bot Jun 19, 2026

Uh oh!

gemini-code-assist Bot Jun 19, 2026

Uh oh!

gemini-code-assist Bot Jun 19, 2026

Uh oh!

gemini-code-assist Bot Jun 19, 2026

Uh oh!

gemini-code-assist Bot Jun 19, 2026

Uh oh!

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

1 participant

	const res = await fetch(`${baseUrl.replace(/\/v1$/, "")}/api/monitoring/health`, {
	const cleanBase = baseUrl.replace(/\/v1\/?$/, "").replace(/\/$/, "");
	const res = await fetch(`${cleanBase}/api/monitoring/health`, {

	const baseUrl = (opts.remote ?? `http://localhost:${port}`).replace(/\/v1$/, "");
	const baseUrl = (opts.remote ?? `http://localhost:${port}`).replace(/\/v1\/?$/, "").replace(/\/$/, "");

	models = body.data ?? body.models ?? [];
	models = Array.isArray(body) ? body : (body?.data ?? body?.models ?? []);

	printError(`Failed to fetch models: ${err.message}`);
	printError(`Failed to fetch models: ${err?.message \|\| String(err)}`);

		import { registerLaunchCodex } from "./launch-codex.mjs";
		import { registerSetupCodex } from "./setup-codex.mjs";

Uh oh!

Conversation

diegosouzapw commented Jun 19, 2026

What

Validation

Uh oh!

chatgpt-codex-connector Bot commented Jun 19, 2026

Uh oh!

gemini-code-assist Bot left a comment

Choose a reason for hiding this comment

Code Review

Uh oh!

gemini-code-assist Bot Jun 19, 2026

Choose a reason for hiding this comment

Uh oh!

gemini-code-assist Bot Jun 19, 2026

Choose a reason for hiding this comment

Uh oh!

gemini-code-assist Bot Jun 19, 2026

Choose a reason for hiding this comment

Uh oh!

gemini-code-assist Bot Jun 19, 2026

Choose a reason for hiding this comment

Uh oh!

gemini-code-assist Bot Jun 19, 2026

Choose a reason for hiding this comment

Uh oh!

gemini-code-assist Bot Jun 19, 2026

Choose a reason for hiding this comment

Uh oh!

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

1 participant