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_feature - Prompt - OpenSmartRoute

Name: create_feature
Author: lhorsager

Prompt file imported from lhorsager/uarm_mcp (.cursor/commands/create_feature.md). Fill in {{arguments}} before use. Copyright stays with the author.

User Input

{{arguments}}

You MUST consider the user input before proceeding (if not empty). The text after the command is the feature description.

Outline

Create a new feature based on the user's description. Use docs/spec/spec.md as the foundation to identify what fits into this feature.

Scripts: Run from repo root. Use bash .specdriven/scripts/bash/create-new-feature.sh on Mac/Linux, or pwsh .specdriven/scripts/powershell/create-new-feature.ps1 on Windows.

Generate short name (2-4 words) from the feature description (e.g., "user-auth", "photo-albums").
Run the create-new-feature script with --json:
- Mac/Linux: bash .specdriven/scripts/bash/create-new-feature.sh --json --short-name "SHORT_NAME" "FEATURE_DESCRIPTION"
- Windows: pwsh .specdriven/scripts/powershell/create-new-feature.ps1 -Json -ShortName "SHORT_NAME" "FEATURE_DESCRIPTION"
- First determine next feature number by checking docs/features/ directories and git branches (pattern ###-*)
- Pass --number N if you calculated the number, or let the script auto-detect
Parse script output for FEATURE_FILE, FEATURE_DIR, BRANCH_NAME.
Load docs/spec/spec.md to understand the solution context. Map the user's feature description to relevant sections of the spec.
Fill feature.md using .specdriven/templates/feature_template.md:
- User Scenarios & Testing (from spec + user input)
- Requirements (feature-specific)
- Success Criteria
- Key Entities
- Dependencies on spec sections
Write to FEATURE_FILE (docs/features/[###-name]/feature.md).
Update docs/spec/features.md to add this feature to the tracker.
Report completion with feature path and next steps (/interrogate_feature, /feature_tech, /build_feature).

Guidelines

Focus on WHAT and WHY, not HOW
Use spec.md to ensure feature aligns with solution
Limit [NEEDS CLARIFICATION] to max 3 critical items

Install create_feature 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.

## User Input ```text {{arguments}} ``` You **MUST** consider the user input before proceeding (if not empty). The text after the command **is** the feature description. ## Outline Create a new feature based on the user's description. Use `docs/spec/spec.md` as the foundation to identify what fits into this feature. **Scripts**: Run from repo root. Use `bash .specdriven/scripts/bash/create-new-feature.sh` on Mac/Linux, or `pwsh .specdriven/scripts/powershell/create-new-feature.ps1` on Windows. 1. **Generate short name** (2-4 words) from the feature description (e.g., "user-auth", "photo-albums"). 2. **Run the create-new-feature script** with `--json`: - **Mac/Linux**: `bash .specdriven/scripts/bash/create-new-feature.sh --json --short-name "SHORT_NAME" "FEATURE_DESCRIPTION"` - **Windows**: `pwsh .specdriven/scripts/powershell/create-new-feature.ps1 -Json -ShortName "SHORT_NAME" "FEATURE_DESCRIPTION"` - First determine next feature number by checking `docs/features/` directories and git branches (pattern `###-*`) - Pass `--number N` if you calculated the number, or let the script auto-detect 3. **Parse script output** for FEATURE_FILE, FEATURE_DIR, BRANCH_NAME. 4. **Load** `docs/spec/spec.md` to understand the solution context. Map the user's feature description to relevant sections of the spec. 5. **Fill feature.md** using `.specdriven/templates/feature_template.md`: - User Scenarios & Testing (from spec + user input) - Requirements (feature-specific) - Success Criteria - Key Entities - Dependencies on spec sections 6. **Write** to FEATURE_FILE (docs/features/[###-name]/feature.md). 7. **Update** `docs/spec/features.md` to add this feature to the tracker. 8. **Report** completion with feature path and next steps (/interrogate_feature, /feature_tech, /build_feature). ## Guidelines - Focus on WHAT and WHY, not HOW - Use spec.md to ensure feature aligns with solution - Limit [NEEDS CLARIFICATION] to max 3 critical items

{ "prompt": "## User Input\n\n```text\n{{arguments}}\n```\n\nYou **MUST** consider the user input before proceeding (if not empty). The text after the command **is** the feature description.\n\n## Outline\n\nCreate a new feature based on the user's description. Use `docs/spec/spec.md` as the foundation to identify what fits into this feature.\n\n**Scripts**: Run from repo root. Use `bash .specdriven/scripts/bash/create-new-feature.sh` on Mac/Linux, or `pwsh .specdriven/scripts/powershell/create-new-feature.ps1` on Windows.\n\n1. **Generate short name** (2-4 words) from the feature description (e.g., \"user-auth\", \"photo-albums\").\n\n2. **Run the create-new-feature script** with `--json`:\n - **Mac/Linux**: `bash .specdriven/scripts/bash/create-new-feature.sh --json --short-name \"SHORT_NAME\" \"FEATURE_DESCRIPTION\"`\n - **Windows**: `pwsh .specdriven/scripts/powershell/create-new-feature.ps1 -Json -ShortName \"SHORT_NAME\" \"FEATURE_DESCRIPTION\"`\n - First determine next feature number by checking `docs/features/` directories and git branches (pattern `###-*`)\n - Pass `--number N` if you calculated the number, or let the script auto-detect\n\n3. **Parse script output** for FEATURE_FILE, FEATURE_DIR, BRANCH_NAME.\n\n4. **Load** `docs/spec/spec.md` to understand the solution context. Map the user's feature description to relevant sections of the spec.\n\n5. **Fill feature.md** using `.specdriven/templates/feature_template.md`:\n - User Scenarios & Testing (from spec + user input)\n - Requirements (feature-specific)\n - Success Criteria\n - Key Entities\n - Dependencies on spec sections\n\n6. **Write** to FEATURE_FILE (docs/features/[###-name]/feature.md).\n\n7. **Update** `docs/spec/features.md` to add this feature to the tracker.\n\n8. **Report** completion with feature path and next steps (/interrogate_feature, /feature_tech, /build_feature).\n\n## Guidelines\n\n- Focus on WHAT and WHY, not HOW\n- Use spec.md to ensure feature aligns with solution\n- Limit [NEEDS CLARIFICATION] to max 3 critical items", "variables": [ "arguments" ], "notes": "Create a new feature. Uses spec.md as foundation to identify what fits. Creates feature folder and feature.md.", "metadata": { "source": { "provider": "github-cursor-commands", "repository": "https://github.com/lhorsager/uarm_mcp", "path": ".cursor/commands/create_feature.md", "ref": "5bd042d94a625ce04bfcf0fa1d83a2ee6289af4d", "url": "https://github.com/lhorsager/uarm_mcp/blob/5bd042d94a625ce04bfcf0fa1d83a2ee6289af4d/.cursor/commands/create_feature.md", "key": "lhorsager/uarm_mcp/.cursor/commands/create_feature.md" } } }

create_feature

About

User Input

Outline

Guidelines

Install create_feature 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