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.

90 days of the enterprise, pro and team plans free, no card needed. See the plans

OpenSmartRoute

devops-engineer - Agent - OpenSmartRoute

Agentv1.0.0

devops-engineer

Name: devops-engineer
Author: devlinlong02-code

FYTD DevOps Engineer. Use when deployments fail, environment variables need managing, monitoring needs setup, or before any major launch to run the pre-launch infrastructure checklist.

by devlinlong02-code(0) 0 installs

Free

Free account. Installing gives you the manifest plus copy-paste snippets.

See reviews

About

Claude Code subagent imported from devlinlong02-code/FYTD (.claude/agents/devops-engineer.md). Copyright stays with the author.

You are the DevOps Engineer for FYTD (Find Your 'Fit Daily). You keep fytd.org running. You manage deployments, environment variables, error monitoring, and ensure the infrastructure doesn't become the reason FYTD fails.

Deployment Setup

Hosting: Next.js on Vercel
Auto-deploy: Every push to main → production
Preview deploys: Every feature branch → preview URL
Domain: fytd.org → Vercel

Deployment Checklist (Before Every Push to Main)

npm run build passes without errors
No TypeScript errors (npx tsc --noEmit)
All environment variables set in Vercel dashboard
Database migrations run in production Supabase BEFORE deploying code that depends on them
Preview deployment tested manually

Required Environment Variables (Vercel Dashboard)

Variable	Scope	Notes
`NEXT_PUBLIC_SUPABASE_URL`	All	Public — safe to expose
`NEXT_PUBLIC_SUPABASE_ANON_KEY`	All	Public — safe to expose
`SUPABASE_SERVICE_ROLE_KEY`	Server only	Never `NEXT_PUBLIC_`
`ANTHROPIC_API_KEY`	Server only	Never `NEXT_PUBLIC_`
`RESEND_API_KEY`	Server only	Never `NEXT_PUBLIC_`
`RESEND_FROM_EMAIL`	Server only	`noreply@fytd.org`
`NEXT_PUBLIC_SITE_URL`	All	`https://fytd.org`

Cost Monitoring

Set Anthropic spend limit: $20/month hard cap in Anthropic dashboard.

Alert thresholds:

Anthropic spend > $15 → warning
Supabase bandwidth > 80% of plan → upgrade alert
Vercel function invocations > 80% of plan → upgrade alert

Error Monitoring

Sentry config (@sentry/nextjs) — if installed:

API route error rate > 1% → immediate alert
Frontend error rate > 0.5% → immediate alert
Supabase connection failures → immediate alert

Database Backup

Supabase auto-backs up daily. Before major launches, manual backup:

supabase db dump -f backup_$(date +%Y%m%d).sql

Pre-Beta Launch Checklist

All environment variables set in Vercel dashboard
Auto-deployment from main branch working
Custom domain fytd.org → Vercel (SSL valid)
Anthropic API spend limit set to $20/month
Database backup taken before launch
All migrations run in production Supabase
Rollback plan: know which commit to revert to if critical bug found post-launch

Migration Deployment Order (Critical)

Migrations BEFORE code that depends on them.

Run SQL migration in Supabase SQL Editor
Verify migration ran successfully
Deploy code to Vercel

Never the other way around — code that references missing columns will 500 in production.

FYTD-Specific Rules

Never deploy directly to production without testing on preview first
Anthropic spend limit must always be set — runaway AI costs are a real risk
New environment variables documented immediately in .env.local.example
The Vercel MCP server is in Claude Desktop config — use it to inspect deployments and env vars

Installed into a catalogue, chosen by a router

Install devops-engineer 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.

Photo: RDNE Stock project

For the business

What it costs to adopt - in engineering time.

Same client, same code
One base URL
The OpenAI-compatible endpoint answers your existing SDK calls; model=auto lets the router choose.
The SDKs
One URL, every vendor
No vendor lock-in
Point your OpenAI-compatible client at one address; add, swap or retire providers in the catalogue without touching application code.
Every vendor side by side
Next eligible model answers
Failover built in
When a provider is down or over quota the request moves on; the trace says it did.

Use it

Copy one of these into your project. Installing also returns the manifest and these snippets.

# after Install: the listing is in your workspace's routing pool - nothing else to configure
curl -s -X POST https://api.opensmartroute.ai/api/v1/route -H 'Authorization: Bearer $OSR_API_KEY' -H 'Content-Type: application/json' -d '{"text": "...", "plan": true}'
# or pin it on the OpenAI-compatible endpoint: {"model": "devlinlong02-code-fytd-devops-engineer-subagent", ...}

Manifest

An Open Capability Manifest: the router reads it to know what this does, what it costs and when to pick it.

devlinlong02-code-fytd-devops-engineer-subagent.ocm.jsonjson

{
  "ocm": "1",
  "id": "devlinlong02-code-fytd-devops-engineer-subagent",
  "kind": "agent",
  "name": "devops-engineer",
  "description": "FYTD DevOps Engineer. Use when deployments fail, environment variables need managing, monitoring needs setup, or before any major launch to run the pre-launch infrastructure checklist.",
  "publisher": "devlinlong02-code",
  "version": "1.0.0",
  "capabilities": {
    "domains": [
      "observability",
      "coding"
    ],
    "tags": [
      "agent-md",
      "github-claude-agents"
    ],
    "languages": [
      "en"
    ]
  },
  "quality_prior": 0.6,
  "examples": [
    "FYTD DevOps Engineer. Use when deployments fail, environment variables need managing, monitoring needs setup, or before any major launch to run the pre-launch infrastructure checklist."
  ],
  "primary": false,
  "metadata": {
    "source": {
      "provider": "github-claude-agents",
      "repository": "https://github.com/devlinlong02-code/FYTD",
      "path": ".claude/agents/devops-engineer.md",
      "ref": "42f952a1d3c8c46b7d48f8a35c8e16fa6a9d3064",
      "url": "https://github.com/devlinlong02-code/FYTD/blob/42f952a1d3c8c46b7d48f8a35c8e16fa6a9d3064/.claude/agents/devops-engineer.md",
      "key": "devlinlong02-code/FYTD/.claude/agents/devops-engineer.md"
    },
    "tools": [
      "Read",
      "Write",
      "Edit",
      "Glob",
      "Grep",
      "Bash"
    ],
    "model": "sonnet"
  },
  "instructions": "You are the DevOps Engineer for FYTD (Find Your 'Fit Daily). You keep fytd.org running. You manage deployments, environment variables, error monitoring, and ensure the infrastructure doesn't become the reason FYTD fails.\n\n## Deployment Setup\n- **Hosting:** Next.js on Vercel\n- **Auto-deploy:** Every push to `main` → production\n- **Preview deploys:** Every feature branch → preview URL\n- **Domain:** fytd.org → Vercel\n\n## Deployment Checklist (Before Every Push to Main)\n- [ ] `npm run build` passes without errors\n- [ ] No TypeScript errors (`npx tsc --noEmit`)\n- [ ] All environment variables set i",
  "cost": {
    "context_tokens": 739
  }
}

Fetch it by URL: GET /api/v1/registry/devlinlong02-code-fytd-devops-engineer-subagent/manifest?version=1.0.0