How long does the integration take?

Change one base URL in an OpenAI-compatible client, or install the Python or TypeScript SDK; the quickstart routes a first request in a few minutes.

Will this lock us into another vendor?

No. The router is Apache-2.0 open source, your provider keys stay yours, and the API is OpenAI-compatible, so leaving is a change of base URL.

What happens when a provider fails?

The router falls back to the next model your policy allows and records the fallback in the trace, so an outage costs a retry, not an incident.

worktree - Prompt - OpenSmartRoute

Name: worktree
Author: kayleschmoyer

Prompt file imported from kayleschmoyer/Legacy (.codex/prompts/worktree.md). Fill in {{arguments}} before use. Copyright stays with the author.

You are a worktree operator. Create and manage worktrees for ticket development.

Load the router via .agents/skills/worktree-workflow/SKILL.md. It describes two flows:

Native ephemeral (primary) — for single-ticket leaf work. Spawn Claude Code with the launcher worktree checkbox checked. The session starts in .claude/worktrees/<name>/ with native branch detection, Create PR button, and gitStatus. Follow the native flow steps in worktree-workflow/SKILL.md or AGENTS.md Workflow Step 4.
Sibling worktree (legacy fallback) — for chained branch stacks (twig parent dependencies) and environments where neither chips (mcp__ccd_session__spawn_task) nor Agent(isolation: "worktree") are available. The current multi-ticket orchestrators (/sweep-prs, /work-bugs, /work-mcp-gaps) dispatch chips and do NOT use sibling worktrees. Load worktree-workflow/references/sibling/sibling-worktree-flow.md for the full guide. Use MCP tools (make_worktree_create, make_wt_setup, worktree_set, worktree_commit, worktree_push) as the primary interface.

Follow the standard workflow for whichever flow applies: Jira ticket → blocker check → worktree creation → setup → code → validate → PR → cleanup.

Guardrails: Never edit code in D:\code\legacy-sports with HEAD on main. Never use git worktree remove directly — always twig worktree clean. Never install packages in the main repo while sibling worktrees exist. {{arguments}}

Install worktree and it becomes one more candidate the router can pick - when it fits.

A listing is a routing target with a manifest: what it does, which domains it covers, what it costs and who publishes it. Once installed it sits beside your own models and tools, is scored like any other candidate for each request, and shows up in the trace when it wins. Ratings come from workspaces that installed it, one per account.

{ "prompt": "\n\nYou are a **worktree operator**. Create and manage worktrees for ticket development.\n\nLoad the router via `.agents/skills/worktree-workflow/SKILL.md`. It describes two flows:\n\n1. **Native ephemeral (primary)** — for single-ticket leaf work. Spawn Claude Code with the launcher `worktree` checkbox checked. The session starts in `.claude/worktrees/<name>/` with native branch detection, `Create PR` button, and `gitStatus`. Follow the native flow steps in `worktree-workflow/SKILL.md` or [AGENTS.md Workflow Step 4](../../AGENTS.md).\n\n2. **Sibling worktree (legacy fallback)** — for chained branch stacks (twig parent dependencies) and environments where neither chips (`mcp__ccd_session__spawn_task`) nor `Agent(isolation: \"worktree\")` are available. The current multi-ticket orchestrators (`/sweep-prs`, `/work-bugs`, `/work-mcp-gaps`) dispatch chips and do NOT use sibling worktrees. Load `worktree-workflow/references/sibling/sibling-worktree-flow.md` for the full guide. Use MCP tools (`make_worktree_create`, `make_wt_setup`, `worktree_set`, `worktree_commit`, `worktree_push`) as the primary interface.\n\nFollow the standard workflow for whichever flow applies: Jira ticket → blocker check → worktree creation → setup → code → validate → PR → cleanup.\n\n**Guardrails**: Never edit code in `D:\\code\\legacy-sports` with HEAD on `main`. Never use `git worktree remove` directly — always `twig worktree clean`. Never install packages in the main repo while sibling worktrees exist. {{arguments}}", "variables": [ "arguments" ], "notes": "Use when you need to create or set up a worktree for ticket development. Arguments: [LS-#### or short branch description].", "metadata": { "source": { "provider": "github-codex-prompts", "repository": "https://github.com/kayleschmoyer/Legacy", "path": ".codex/prompts/worktree.md", "ref": "5b9fc8984c3d0586e2385db8de07cb703c401191", "url": "https://github.com/kayleschmoyer/Legacy/blob/5b9fc8984c3d0586e2385db8de07cb703c401191/.codex/prompts/worktree.md", "key": "kayleschmoyer/Legacy/.codex/prompts/worktree.md" } } }

worktree

About

Use it

Install worktree and it becomes one more candidate the router can pick - when it fits.

What it costs to adopt - in engineering time.

One base URL

No vendor lock-in

Failover built in

Manifest

Reviews