diff --git a/plugins/browser/firecrawl/provider.py b/plugins/browser/firecrawl/provider.py
index 2c605134a01..a2bfee60ba8 100644
--- a/plugins/browser/firecrawl/provider.py
+++ b/plugins/browser/firecrawl/provider.py
@@ -39,6 +39,12 @@ logger = logging.getLogger(__name__)
 
 _BASE_URL = "https://api.firecrawl.dev"
 
+# Integration tag sent on every browser-session create so Firecrawl can
+# attribute Hermes usage. Mirrors FIRECRAWL_INTEGRATION_TAG in the web
+# plugin; kept here rather than imported to keep the two firecrawl
+# plugins import-independent.
+FIRECRAWL_INTEGRATION_TAG = "hermes"
+
 
 class FirecrawlBrowserProvider(BrowserProvider):
     """Firecrawl (https://firecrawl.dev) cloud browser backend.
@@ -80,7 +86,10 @@ class FirecrawlBrowserProvider(BrowserProvider):
     def create_session(self, task_id: str) -> Dict[str, object]:
         ttl = int(os.environ.get("FIRECRAWL_BROWSER_TTL", "300"))
 
-        body: Dict[str, object] = {"ttl": ttl}
+        body: Dict[str, object] = {
+            "ttl": ttl,
+            "integration": FIRECRAWL_INTEGRATION_TAG,
+        }
 
         try:
             response = requests.post(
diff --git a/plugins/web/firecrawl/provider.py b/plugins/web/firecrawl/provider.py
index bcc574ffca3..f57a6113185 100644
--- a/plugins/web/firecrawl/provider.py
+++ b/plugins/web/firecrawl/provider.py
@@ -56,6 +56,11 @@ from tools.website_policy import check_website_access
 logger = logging.getLogger(__name__)
 
 
+# Integration tag passed to Firecrawl on every call so they can attribute
+# usage back to Hermes.
+FIRECRAWL_INTEGRATION_TAG = "hermes"
+
+
 # ---------------------------------------------------------------------------
 # Lazy Firecrawl SDK proxy
 # ---------------------------------------------------------------------------
@@ -408,7 +413,11 @@ class FirecrawlWebSearchProvider(WebSearchProvider):
         # let it propagate so the dispatcher emits the legacy envelope shape.
         client = _get_firecrawl_client()
         try:
-            response = client.search(query=query, limit=limit)
+            response = client.search(
+                query=query,
+                limit=limit,
+                integration=FIRECRAWL_INTEGRATION_TAG,
+            )
             web_results = _extract_web_search_results(response)
             logger.info("Firecrawl: found %d search results", len(web_results))
             return {"success": True, "data": {"web": web_results}}
@@ -487,6 +496,7 @@ class FirecrawlWebSearchProvider(WebSearchProvider):
                             _get_firecrawl_client().scrape,
                             url=url,
                             formats=formats,
+                            integration=FIRECRAWL_INTEGRATION_TAG,
                         ),
                         timeout=60,
                     )
@@ -623,6 +633,7 @@ class FirecrawlWebSearchProvider(WebSearchProvider):
             crawl_params = {
                 "limit": limit,
                 "scrape_options": {"formats": ["markdown"]},
+                "integration": FIRECRAWL_INTEGRATION_TAG,
             }
 
             # The SDK call is sync; run in a thread so we don't block the