79 lines
2.8 KiB
Python
79 lines
2.8 KiB
Python
"""Firecrawl cloud browser provider (plugin form).
|
|
|
|
Cloud-browser path only (``/v2/browser``) — distinct from the firecrawl WEB
|
|
plugin at ``plugins/web/firecrawl/`` (search/extract/crawl on ``/v2/*``); both
|
|
share ``FIRECRAWL_API_KEY``.
|
|
|
|
Config: ``browser.cloud_provider: "firecrawl"`` (explicit selection only — not
|
|
in the legacy auto-detect walk). Env: ``FIRECRAWL_API_KEY``
|
|
(https://firecrawl.dev), ``FIRECRAWL_API_URL`` (default
|
|
https://api.firecrawl.dev), ``FIRECRAWL_BROWSER_TTL`` (default 300 seconds).
|
|
"""
|
|
|
|
from __future__ import annotations
|
|
|
|
import logging
|
|
import os
|
|
from typing import Any, Dict, Optional
|
|
|
|
from agent.secret_scope import get_secret
|
|
from plugins.browser._common import CloudBrowserProvider
|
|
|
|
logger = logging.getLogger(__name__)
|
|
|
|
_BASE_URL = "https://api.firecrawl.dev"
|
|
|
|
|
|
class FirecrawlBrowserProvider(CloudBrowserProvider):
|
|
"""Firecrawl (https://firecrawl.dev) cloud browser backend."""
|
|
|
|
provider_id = "firecrawl"
|
|
label = "Firecrawl"
|
|
release_method = "delete"
|
|
release_path = "/v2/browser/{session_id}"
|
|
create_label_suffix = " browser"
|
|
setup_tag = "Cloud browser with remote execution"
|
|
setup_env_vars = [
|
|
{"key": "FIRECRAWL_API_KEY", "prompt": "Firecrawl API key", "url": "https://firecrawl.dev"},
|
|
]
|
|
|
|
def _api_url(self) -> str:
|
|
return os.environ.get("FIRECRAWL_API_URL", _BASE_URL)
|
|
|
|
def _get_config_or_none(self) -> Optional[Dict[str, Any]]:
|
|
return {"base_url": self._api_url()} if get_secret("FIRECRAWL_API_KEY") else None
|
|
|
|
def _get_config(self) -> Dict[str, Any]:
|
|
# Never raises: a missing key surfaces from _headers() inside the request
|
|
# try-block, so close_session logs it as an exception (legacy behaviour).
|
|
return {"base_url": self._api_url()}
|
|
|
|
def _headers(self, config: Optional[Dict[str, Any]] = None) -> Dict[str, str]:
|
|
api_key = get_secret("FIRECRAWL_API_KEY")
|
|
if not api_key:
|
|
raise ValueError(
|
|
"FIRECRAWL_API_KEY environment variable is required. "
|
|
"Get your key at https://firecrawl.dev"
|
|
)
|
|
return {"Content-Type": "application/json", "Authorization": f"Bearer {api_key}"}
|
|
|
|
def create_session(self, task_id: str) -> Dict[str, object]:
|
|
try:
|
|
ttl = int(os.environ.get("FIRECRAWL_BROWSER_TTL", "300"))
|
|
except (ValueError, TypeError):
|
|
ttl = 300
|
|
|
|
response = self._post_create(f"{self._api_url()}/v2/browser", self._headers(), {"ttl": ttl})
|
|
self._check_created(response)
|
|
|
|
data = response.json()
|
|
session_name = self._session_name(task_id)
|
|
logger.info("Created Firecrawl browser session %s", session_name)
|
|
|
|
return {
|
|
"session_name": session_name,
|
|
"bb_session_id": data["id"],
|
|
"cdp_url": data["cdpUrl"],
|
|
"features": {"firecrawl": True},
|
|
}
|