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.

create_worktree - Prompt - OpenSmartRoute

Name: create_worktree
Author: itissid

Prompt file imported from itissid/wiki (.claude/commands/create_worktree.md). Fill in {{HUMANLAYER_WORKTREE_OVERRIDE_BASE}} before use. Copyright stays with the author.

description: Create worktree and launch implementation session for a plan

determine worktree base path: 1a. Ask the user if they want to use a custom worktree base path via HUMANLAYER_WORKTREE_OVERRIDE_BASE
- Default is ~/wt if not set
- If user provides a custom path, export it before running create_worktree.sh: export HUMANLAYER_WORKTREE_OVERRIDE_BASE=/custom/path
- The create_worktree.sh script will use $HUMANLAYER_WORKTREE_OVERRIDE_BASE/$REPO_NAME/ as the base
set up worktree for implementation: 2a. If $REPO_NAME is humanlayer you should read from hack/create_worktree.sh if not read create_worktree.sh (i.e. it should be availaible via the PATH) and create a new worktree with the Linear branch name using : ./hack/create_worktree.sh ENG-XXXX BRANCH_NAME OR create_worktree.sh ENG-XXXX BRANCH_NAME. Accordingly pass the --no-setup option if the repo is not humanlayer. If create_worktree.sh is not available halt and tell the user you cannot find this executable and cannot go on.
determine required data:

branch name path to plan file (use relative path only) launch prompt command to run

IMPORTANT PATH USAGE:

The thoughts/ directory is synced between the main repo and worktrees
Always use ONLY the relative path starting with thoughts/shared/... without any directory prefix
Example: thoughts/shared/plans/fix-mcp-keepalive-proper.md (not the full absolute path)
This works because thoughts are synced and accessible from the worktree

3a. confirm with the user by sending a message to the Human

based on the input, I plan to create a worktree with the following details:

worktree base: {{HUMANLAYER_WORKTREE_OVERRIDE_BASE}}
worktree path: {{HUMANLAYER_WORKTREE_OVERRIDE_BASE}}/$REPO_NAME/ENG-XXXX
branch name: BRANCH_NAME
path to plan file: $FILEPATH
launch prompt:

    /implement_plan at $FILEPATH and when you are done implementing and all tests pass, read ./claude/commands/commit.md and create a commit, then read ./claude/commands/describe_pr.md and create a PR, then add a comment to the Linear ticket with the PR link

command to run:

    humanlayer launch --model opus -w {{HUMANLAYER_WORKTREE_OVERRIDE_BASE}}/$REPO_NAME/ENG-XXXX "/implement_plan at $FILEPATH and when you are done implementing and all tests pass, read ./claude/commands/commit.md and create a commit, then read ./claude/commands/describe_pr.md and create a PR, then add a comment to the Linear ticket with the PR link"

incorporate any user feedback then:

launch implementation session: humanlayer launch --model opus -w {{HUMANLAYER_WORKTREE_OVERRIDE_BASE}}/$REPO_NAME/ENG-XXXX "/implement_plan at $FILEPATH and when you are done implementing and all tests pass, read ./claude/commands/commit.md and create a commit, then read ./claude/commands/describe_pr.md and create a PR, then add a comment to the Linear ticket with the PR link"

Install create_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.

--- description: Create worktree and launch implementation session for a plan --- 1. determine worktree base path: 1a. Ask the user if they want to use a custom worktree base path via `HUMANLAYER_WORKTREE_OVERRIDE_BASE` - Default is `~/wt` if not set - If user provides a custom path, export it before running create_worktree.sh: `export HUMANLAYER_WORKTREE_OVERRIDE_BASE=/custom/path` - The create_worktree.sh script will use `$HUMANLAYER_WORKTREE_OVERRIDE_BASE/$REPO_NAME/` as the base 2. set up worktree for implementation: 2a. If $REPO_NAME is `humanlayer` you should read from `hack/create_worktree.sh` if not read create_worktree.sh (i.e. it should be availaible via the PATH) and create a new worktree with the Linear branch name using : `./hack/create_worktree.sh ENG-XXXX BRANCH_NAME` OR `create_worktree.sh ENG-XXXX BRANCH_NAME`. Accordingly pass the --no-setup option if the repo is not humanlayer. If create_worktree.sh is not available halt and tell the user you cannot find this executable and cannot go on. 3. determine required data: branch name path to plan file (use relative path only) launch prompt command to run **IMPORTANT PATH USAGE:** - The thoughts/ directory is synced between the main repo and worktrees - Always use ONLY the relative path starting with `thoughts/shared/...` without any directory prefix - Example: `thoughts/shared/plans/fix-mcp-keepalive-proper.md` (not the full absolute path) - This works because thoughts are synced and accessible from the worktree 3a. confirm with the user by sending a message to the Human ``` based on the input, I plan to create a worktree with the following details: worktree base: {{HUMANLAYER_WORKTREE_OVERRIDE_BASE}} worktree path: {{HUMANLAYER_WORKTREE_OVERRIDE_BASE}}/$REPO_NAME/ENG-XXXX branch name: BRANCH_NAME path to plan file: $FILEPATH launch prompt: /implement_plan at $FILEPATH and when you are done implementing and all tests pass, read ./claude/commands/commit.md and create a commit, then read ./claude/commands/describe_pr.md and create a PR, then add a comment to the Linear ticket with the PR link command to run: humanlayer launch --model opus -w {{HUMANLAYER_WORKTREE_OVERRIDE_BASE}}/$REPO_NAME/ENG-XXXX "/implement_plan at $FILEPATH and when you are done implementing and all tests pass, read ./claude/commands/commit.md and create a commit, then read ./claude/commands/describe_pr.md and create a PR, then add a comment to the Linear ticket with the PR link" ``` incorporate any user feedback then: 4. launch implementation session: `humanlayer launch --model opus -w {{HUMANLAYER_WORKTREE_OVERRIDE_BASE}}/$REPO_NAME/ENG-XXXX "/implement_plan at $FILEPATH and when you are done implementing and all tests pass, read ./claude/commands/commit.md and create a commit, then read ./claude/commands/describe_pr.md and create a PR, then add a comment to the Linear ticket with the PR link"`

{ "prompt": "---\ndescription: Create worktree and launch implementation session for a plan\n---\n\n1. determine worktree base path:\n1a. Ask the user if they want to use a custom worktree base path via `HUMANLAYER_WORKTREE_OVERRIDE_BASE`\n - Default is `~/wt` if not set\n - If user provides a custom path, export it before running create_worktree.sh: `export HUMANLAYER_WORKTREE_OVERRIDE_BASE=/custom/path`\n - The create_worktree.sh script will use `$HUMANLAYER_WORKTREE_OVERRIDE_BASE/$REPO_NAME/` as the base\n\n2. set up worktree for implementation:\n2a. If $REPO_NAME is `humanlayer` you should read from `hack/create_worktree.sh` if not read create_worktree.sh (i.e. it should be availaible via the PATH) and create a new worktree with the Linear branch name using : `./hack/create_worktree.sh ENG-XXXX BRANCH_NAME` OR `create_worktree.sh ENG-XXXX BRANCH_NAME`. Accordingly pass the --no-setup option if the repo is not humanlayer. If create_worktree.sh is not available halt and tell the user you cannot find this executable and cannot go on.\n\n3. determine required data:\n\nbranch name\npath to plan file (use relative path only)\nlaunch prompt\ncommand to run\n\n**IMPORTANT PATH USAGE:**\n- The thoughts/ directory is synced between the main repo and worktrees\n- Always use ONLY the relative path starting with `thoughts/shared/...` without any directory prefix\n- Example: `thoughts/shared/plans/fix-mcp-keepalive-proper.md` (not the full absolute path)\n- This works because thoughts are synced and accessible from the worktree\n\n3a. confirm with the user by sending a message to the Human\n\n```\nbased on the input, I plan to create a worktree with the following details:\n\nworktree base: {{HUMANLAYER_WORKTREE_OVERRIDE_BASE}}\nworktree path: {{HUMANLAYER_WORKTREE_OVERRIDE_BASE}}/$REPO_NAME/ENG-XXXX\nbranch name: BRANCH_NAME\npath to plan file: $FILEPATH\nlaunch prompt:\n\n /implement_plan at $FILEPATH and when you are done implementing and all tests pass, read ./claude/commands/commit.md and create a commit, then read ./claude/commands/describe_pr.md and create a PR, then add a comment to the Linear ticket with the PR link\n\ncommand to run:\n\n humanlayer launch --model opus -w {{HUMANLAYER_WORKTREE_OVERRIDE_BASE}}/$REPO_NAME/ENG-XXXX \"/implement_plan at $FILEPATH and when you are done implementing and all tests pass, read ./claude/commands/commit.md and create a commit, then read ./claude/commands/describe_pr.md and create a PR, then add a comment to the Linear ticket with the PR link\"\n```\n\nincorporate any user feedback then:\n\n4. launch implementation session: `humanlayer launch --model opus -w {{HUMANLAYER_WORKTREE_OVERRIDE_BASE}}/$REPO_NAME/ENG-XXXX \"/implement_plan at $FILEPATH and when you are done implementing and all tests pass, read ./claude/commands/commit.md and create a commit, then read ./claude/commands/describe_pr.md and create a PR, then add a comment to the Linear ticket with the PR link\"`", "variables": [ "HUMANLAYER_WORKTREE_OVERRIDE_BASE" ], "notes": "1. determine worktree base path: 1a. Ask the user if they want to use a custom worktree base path via `HUMANLAYER_WORKTREE_OVERRIDE_BASE` - Default is `~/wt` if not set - If user provides a custom path, export it before running create_worktree.sh: `export HUMANLAYER_WORKTREE_OVERRIDE_BASE=/custom/path` - The create_worktree.sh script will use `$HUMANLAYER_WORKTREE_OVERRIDE_BASE/$REPO_NAME/` as the base", "metadata": { "source": { "provider": "github-claude-commands", "repository": "https://github.com/itissid/wiki", "path": ".claude/commands/create_worktree.md", "ref": "410a3c91499367e31c675e5547529974b3f909b7", "url": "https://github.com/itissid/wiki/blob/410a3c91499367e31c675e5547529974b3f909b7/.claude/commands/create_worktree.md", "key": "itissid/wiki/.claude/commands/create_worktree.md" } } }

create_worktree

About

description: Create worktree and launch implementation session for a plan

Install create_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

Use it

Manifest

Reviews