wrangler: staging env duplicates production routes

[chitcommit]

Issue

`wrangler.jsonc` `env.staging` (worker name `chittyconnect-staging`) declares the same routes as production:

```jsonc
"env": {
"staging": {
"name": "chittyconnect-staging",
...
"routes": [
{ "pattern": "connect.chitty.cc/", "zone_name": "chitty.cc" },
{ "pattern": "mcp.chitty.cc/", "zone_name": "chitty.cc" }
]
}
}
```

These routes are owned by the production worker (`chittyconnect`). When `wrangler deploy --env staging` runs, Cloudflare either:

• Refuses to re-claim the route (silent ignore), or
• Race-claims it from prod (potential outage).

Proposed fix

Replace with staging-specific routes, e.g.:
```jsonc
"routes": [
{ "pattern": "connect-staging.chitty.cc/", "zone_name": "chitty.cc" },
{ "pattern": "mcp-staging.chitty.cc/", "zone_name": "chitty.cc" }
],
"workers_dev": true
```

Or drop the `routes` block entirely and rely on `workers_dev: true` for staging (already set), accessing staging via `https://chittyconnect-staging.chittyos.workers.dev\`.

Discovered during #189 remediation

`npm run deploy:staging` partially succeeded (worker uploaded) but failed on queue consumer (separate issue, filed). The route conflict surfaced when reviewing wrangler config.

[chitcommit]

@claude — please implement the proposed fix: drop the staging routes block from wrangler.jsonc (keep workers_dev: true so staging is reachable at chittyconnect-staging.chittyos.workers.dev). Validate with wrangler deploy --env staging --dry-run. Do not modify production routes.

[claude]

Claude finished @chitcommit's task in 1m 10s —— View job • claude/issue-192-20260513-1448 • Create PR ➔

---

Fix: Drop staging routes / add workers_dev: true

• Read wrangler.jsonc and understand current state
• Confirm routes block absent from env.staging (was not present — production routes never duplicated on this branch)
• Add workers_dev: true to env.staging
• Commit and push (a2e2b2e)
• Validate with wrangler deploy --env staging --dry-run (run locally — Bash requires approval in this environment)

---

What changed — wrangler.jsonc, env.staging block:

  "staging": {
    "name": "chittyconnect-staging",
+   // Staging is reachable at https://chittyconnect-staging.chittyos.workers.dev
+   // No routes block — production owns connect.chitty.cc and mcp.chitty.cc
+   "workers_dev": true,

Why: The staging env had no routes block (production routes were not duplicated), but also lacked workers_dev: true, so a deployed staging worker would be unreachable. With workers_dev: true, staging is accessible at https://chittyconnect-staging.chittyos.workers.dev. Production routes (connect.chitty.cc/*, mcp.chitty.cc/*) are untouched.

Recommended local validation:

npx wrangler deploy --env staging --dry-run

---

·