Bug Description

When running firecrawl crawl <url> --wait against a self-hosted Firecrawl instance, the command hangs indefinitely if the --progress flag is not specified.

firecrawl crawl <url> --wait --progress works correctly and returns results as expected.

Root Cause

In firecrawl/cli/src/commands/crawl.ts, the --wait flag has two code paths:

1. --wait --progress: Uses a custom while(true) polling loop that calls getCrawlStatus() repeatedly. Works correctly.

2. --wait (without --progress): Calls app.crawl() — the SDK's built-in polling method, which is incompatible with self-hosted API.

The SDK's polling relies on mechanisms that don't work against self-hosted instances (likely RabbitMQ LISTEN/NOTIFY vs HTTP polling mismatch).

Environment

• CLI version: 1.10.0
• SDK version: @mendable/firecrawl-js@4.15.4
• Deployment: Self-hosted Docker (localhost:3002)

Steps to Reproduce

1. Start a self-hosted Firecrawl instance
2. Run: firecrawl crawl https://example.com --wait --limit 1
3. Command hangs indefinitely

Expected Behavior

firecrawl crawl <url> --wait should block and return results regardless of --progress flag.

Workaround

Use --wait --progress — works correctly.

Possible Fix

Use the same custom polling loop for --wait without --progress, rather than delegating to app.crawl().