diff --git a/skills/productivity/google-workspace/SKILL.md b/skills/productivity/google-workspace/SKILL.md
index 00d91de909e..453662540f6 100644
--- a/skills/productivity/google-workspace/SKILL.md
+++ b/skills/productivity/google-workspace/SKILL.md
@@ -1,6 +1,6 @@
 ---
 name: google-workspace
-description: Gmail, Calendar, Drive, Contacts, Sheets, and Docs integration via Python. Uses OAuth2 with automatic token refresh. No external binaries needed — runs entirely with Google's Python client libraries in the Hermes venv.
+description: Gmail, Calendar, Drive, Contacts, Sheets, and Docs integration for Hermes. Uses Hermes-managed OAuth2 setup, prefers the Google Workspace CLI (`gws`) when available for broader API coverage, and falls back to the Python client libraries otherwise.
 version: 1.0.0
 author: Nous Research
 license: MIT
@@ -13,7 +13,7 @@ metadata:
 
 # Google Workspace
 
-Gmail, Calendar, Drive, Contacts, Sheets, and Docs — all through Python scripts in this skill. No external binaries to install.
+Gmail, Calendar, Drive, Contacts, Sheets, and Docs — through Hermes-managed OAuth and a thin CLI wrapper. When `gws` is installed, the skill uses it as the execution backend for broader Google Workspace coverage; otherwise it falls back to the bundled Python client implementation.
 
 ## References
 
@@ -22,7 +22,7 @@ Gmail, Calendar, Drive, Contacts, Sheets, and Docs — all through Python script
 ## Scripts
 
 - `scripts/setup.py` — OAuth2 setup (run once to authorize)
-- `scripts/google_api.py` — API wrapper CLI (agent uses this for all operations)
+- `scripts/google_api.py` — compatibility wrapper CLI. It prefers `gws` for operations when available, while preserving Hermes' existing JSON output contract.
 
 ## First-Time Setup
 
@@ -122,6 +122,7 @@ Should print `AUTHENTICATED`. Setup is complete — token refreshes automaticall
 
 - Token is stored at `~/.hermes/google_token.json` and auto-refreshes.
 - Pending OAuth session state/verifier are stored temporarily at `~/.hermes/google_oauth_pending.json` until exchange completes.
+- If `gws` is installed, `google_api.py` points it at the same `~/.hermes/google_token.json` credentials file. Users do not need to run a separate `gws auth login` flow.
 - To revoke: `$GSETUP --revoke`
 
 ## Usage
diff --git a/skills/productivity/google-workspace/scripts/google_api.py b/skills/productivity/google-workspace/scripts/google_api.py
index 19c1159d264..13affb849d0 100644
--- a/skills/productivity/google-workspace/scripts/google_api.py
+++ b/skills/productivity/google-workspace/scripts/google_api.py
@@ -1,8 +1,9 @@
 #!/usr/bin/env python3
 """Google Workspace API CLI for Hermes Agent.
 
-A thin CLI wrapper around Google's Python client libraries.
-Authenticates using the token stored by setup.py.
+Uses the Google Workspace CLI (`gws`) when available, but preserves the
+existing Hermes-facing JSON contract and falls back to the Python client
+libraries if `gws` is not installed.
 
 Usage:
   python google_api.py gmail search "is:unread" [--max 10]
@@ -23,6 +24,8 @@ import argparse
 import base64
 import json
 import os
+import shutil
+import subprocess
 import sys
 from datetime import datetime, timedelta, timezone
 from email.mime.text import MIMEText
@@ -30,6 +33,7 @@ from pathlib import Path
 
 HERMES_HOME = Path(os.getenv("HERMES_HOME", Path.home() / ".hermes"))
 TOKEN_PATH = HERMES_HOME / "google_token.json"
+CLIENT_SECRET_PATH = HERMES_HOME / "google_client_secret.json"
 
 SCOPES = [
     "https://www.googleapis.com/auth/gmail.readonly",
@@ -43,13 +47,112 @@ SCOPES = [
 ]
 
 
-def get_credentials():
-    """Load and refresh credentials from token file."""
+def _ensure_authenticated():
     if not TOKEN_PATH.exists():
         print("Not authenticated. Run the setup script first:", file=sys.stderr)
         print(f"  python {Path(__file__).parent / 'setup.py'}", file=sys.stderr)
         sys.exit(1)
 
+
+def _gws_binary() -> str | None:
+    override = os.getenv("HERMES_GWS_BIN")
+    if override:
+        return override
+    return shutil.which("gws")
+
+
+def _gws_env() -> dict[str, str]:
+    env = os.environ.copy()
+    env["GOOGLE_WORKSPACE_CLI_CREDENTIALS_FILE"] = str(TOKEN_PATH)
+    return env
+
+
+def _run_gws(parts: list[str], *, params: dict | None = None, body: dict | None = None):
+    binary = _gws_binary()
+    if not binary:
+        raise RuntimeError("gws not installed")
+
+    _ensure_authenticated()
+
+    cmd = [binary, *parts]
+    if params is not None:
+        cmd.extend(["--params", json.dumps(params)])
+    if body is not None:
+        cmd.extend(["--json", json.dumps(body)])
+
+    result = subprocess.run(
+        cmd,
+        capture_output=True,
+        text=True,
+        env=_gws_env(),
+    )
+    if result.returncode != 0:
+        err = result.stderr.strip() or result.stdout.strip() or "Unknown gws error"
+        print(err, file=sys.stderr)
+        sys.exit(result.returncode or 1)
+
+    stdout = result.stdout.strip()
+    if not stdout:
+        return {}
+
+    try:
+        return json.loads(stdout)
+    except json.JSONDecodeError:
+        print("ERROR: Unexpected non-JSON output from gws:", file=sys.stderr)
+        print(stdout, file=sys.stderr)
+        sys.exit(1)
+
+
+def _headers_dict(msg: dict) -> dict[str, str]:
+    return {h["name"]: h["value"] for h in msg.get("payload", {}).get("headers", [])}
+
+
+def _extract_message_body(msg: dict) -> str:
+    body = ""
+    payload = msg.get("payload", {})
+    if payload.get("body", {}).get("data"):
+        body = base64.urlsafe_b64decode(payload["body"]["data"]).decode("utf-8", errors="replace")
+    elif payload.get("parts"):
+        for part in payload["parts"]:
+            if part.get("mimeType") == "text/plain" and part.get("body", {}).get("data"):
+                body = base64.urlsafe_b64decode(part["body"]["data"]).decode("utf-8", errors="replace")
+                break
+        if not body:
+            for part in payload["parts"]:
+                if part.get("mimeType") == "text/html" and part.get("body", {}).get("data"):
+                    body = base64.urlsafe_b64decode(part["body"]["data"]).decode("utf-8", errors="replace")
+                    break
+    return body
+
+
+def _extract_doc_text(doc: dict) -> str:
+    text_parts = []
+    for element in doc.get("body", {}).get("content", []):
+        paragraph = element.get("paragraph", {})
+        for pe in paragraph.get("elements", []):
+            text_run = pe.get("textRun", {})
+            if text_run.get("content"):
+                text_parts.append(text_run["content"])
+    return "".join(text_parts)
+
+
+def _datetime_with_timezone(value: str) -> str:
+    if not value:
+        return value
+    if "T" not in value:
+        return value
+    if value.endswith("Z"):
+        return value
+    tail = value[10:]
+    if "+" in tail or "-" in tail:
+        return value
+    return value + "Z"
+
+
+def get_credentials():
+    """Load and refresh credentials from token file."""
+    _ensure_authenticated()
+
     from google.oauth2.credentials import Credentials
     from google.auth.transport.requests import Request
 
@@ -65,6 +168,7 @@ def get_credentials():
 
 def build_service(api, version):
     from googleapiclient.discovery import build
+
     return build(api, version, credentials=get_credentials())
 
 
@@ -72,7 +176,41 @@ def build_service(api, version):
 # Gmail
 # =========================================================================
 
+
 def gmail_search(args):
+    if _gws_binary():
+        results = _run_gws(
+            ["gmail", "users", "messages", "list"],
+            params={"userId": "me", "q": args.query, "maxResults": args.max},
+        )
+        messages = results.get("messages", [])
+        output = []
+        for msg_meta in messages:
+            msg = _run_gws(
+                ["gmail", "users", "messages", "get"],
+                params={
+                    "userId": "me",
+                    "id": msg_meta["id"],
+                    "format": "metadata",
+                    "metadataHeaders": ["From", "To", "Subject", "Date"],
+                },
+            )
+            headers = _headers_dict(msg)
+            output.append(
+                {
+                    "id": msg["id"],
+                    "threadId": msg["threadId"],
+                    "from": headers.get("From", ""),
+                    "to": headers.get("To", ""),
+                    "subject": headers.get("Subject", ""),
+                    "date": headers.get("Date", ""),
+                    "snippet": msg.get("snippet", ""),
+                    "labels": msg.get("labelIds", []),
+                }
+            )
+        print(json.dumps(output, indent=2, ensure_ascii=False))
+        return
+
     service = build_service("gmail", "v1")
     results = service.users().messages().list(
         userId="me", q=args.query, maxResults=args.max
@@ -88,7 +226,7 @@ def gmail_search(args):
             userId="me", id=msg_meta["id"], format="metadata",
             metadataHeaders=["From", "To", "Subject", "Date"],
         ).execute()
-        headers = {h["name"]: h["value"] for h in msg.get("payload", {}).get("headers", [])}
+        headers = _headers_dict(msg)
         output.append({
             "id": msg["id"],
             "threadId": msg["threadId"],
@@ -102,30 +240,33 @@ def gmail_search(args):
     print(json.dumps(output, indent=2, ensure_ascii=False))
 
 
+
 def gmail_get(args):
+    if _gws_binary():
+        msg = _run_gws(
+            ["gmail", "users", "messages", "get"],
+            params={"userId": "me", "id": args.message_id, "format": "full"},
+        )
+        headers = _headers_dict(msg)
+        result = {
+            "id": msg["id"],
+            "threadId": msg["threadId"],
+            "from": headers.get("From", ""),
+            "to": headers.get("To", ""),
+            "subject": headers.get("Subject", ""),
+            "date": headers.get("Date", ""),
+            "labels": msg.get("labelIds", []),
+            "body": _extract_message_body(msg),
+        }
+        print(json.dumps(result, indent=2, ensure_ascii=False))
+        return
+
     service = build_service("gmail", "v1")
     msg = service.users().messages().get(
         userId="me", id=args.message_id, format="full"
     ).execute()
 
-    headers = {h["name"]: h["value"] for h in msg.get("payload", {}).get("headers", [])}
-
-    # Extract body text
-    body = ""
-    payload = msg.get("payload", {})
-    if payload.get("body", {}).get("data"):
-        body = base64.urlsafe_b64decode(payload["body"]["data"]).decode("utf-8", errors="replace")
-    elif payload.get("parts"):
-        for part in payload["parts"]:
-            if part.get("mimeType") == "text/plain" and part.get("body", {}).get("data"):
-                body = base64.urlsafe_b64decode(part["body"]["data"]).decode("utf-8", errors="replace")
-                break
-        if not body:
-            for part in payload["parts"]:
-                if part.get("mimeType") == "text/html" and part.get("body", {}).get("data"):
-                    body = base64.urlsafe_b64decode(part["body"]["data"]).decode("utf-8", errors="replace")
-                    break
-
+    headers = _headers_dict(msg)
     result = {
         "id": msg["id"],
         "threadId": msg["threadId"],
@@ -134,12 +275,33 @@ def gmail_get(args):
         "subject": headers.get("Subject", ""),
         "date": headers.get("Date", ""),
         "labels": msg.get("labelIds", []),
-        "body": body,
+        "body": _extract_message_body(msg),
     }
     print(json.dumps(result, indent=2, ensure_ascii=False))
 
 
+
 def gmail_send(args):
+    if _gws_binary():
+        message = MIMEText(args.body, "html" if args.html else "plain")
+        message["to"] = args.to
+        message["subject"] = args.subject
+        if args.cc:
+            message["cc"] = args.cc
+
+        raw = base64.urlsafe_b64encode(message.as_bytes()).decode()
+        body = {"raw": raw}
+        if args.thread_id:
+            body["threadId"] = args.thread_id
+
+        result = _run_gws(
+            ["gmail", "users", "messages", "send"],
+            params={"userId": "me"},
+            body=body,
+        )
+        print(json.dumps({"status": "sent", "id": result["id"], "threadId": result.get("threadId", "")}, indent=2))
+        return
+
     service = build_service("gmail", "v1")
     message = MIMEText(args.body, "html" if args.html else "plain")
     message["to"] = args.to
@@ -157,14 +319,46 @@ def gmail_send(args):
     print(json.dumps({"status": "sent", "id": result["id"], "threadId": result.get("threadId", "")}, indent=2))
 
 
+
 def gmail_reply(args):
+    if _gws_binary():
+        original = _run_gws(
+            ["gmail", "users", "messages", "get"],
+            params={
+                "userId": "me",
+                "id": args.message_id,
+                "format": "metadata",
+                "metadataHeaders": ["From", "Subject", "Message-ID"],
+            },
+        )
+        headers = _headers_dict(original)
+
+        subject = headers.get("Subject", "")
+        if not subject.startswith("Re:"):
+            subject = f"Re: {subject}"
+
+        message = MIMEText(args.body)
+        message["to"] = headers.get("From", "")
+        message["subject"] = subject
+        if headers.get("Message-ID"):
+            message["In-Reply-To"] = headers["Message-ID"]
+            message["References"] = headers["Message-ID"]
+
+        raw = base64.urlsafe_b64encode(message.as_bytes()).decode()
+        result = _run_gws(
+            ["gmail", "users", "messages", "send"],
+            params={"userId": "me"},
+            body={"raw": raw, "threadId": original["threadId"]},
+        )
+        print(json.dumps({"status": "sent", "id": result["id"], "threadId": result.get("threadId", "")}, indent=2))
+        return
+
     service = build_service("gmail", "v1")
-    # Fetch original to get thread ID and headers
     original = service.users().messages().get(
         userId="me", id=args.message_id, format="metadata",
         metadataHeaders=["From", "Subject", "Message-ID"],
     ).execute()
-    headers = {h["name"]: h["value"] for h in original.get("payload", {}).get("headers", [])}
+    headers = _headers_dict(original)
 
     subject = headers.get("Subject", "")
     if not subject.startswith("Re:"):
@@ -184,20 +378,38 @@ def gmail_reply(args):
     print(json.dumps({"status": "sent", "id": result["id"], "threadId": result.get("threadId", "")}, indent=2))
 
 
+
 def gmail_labels(args):
+    if _gws_binary():
+        results = _run_gws(["gmail", "users", "labels", "list"], params={"userId": "me"})
+        labels = [{"id": l["id"], "name": l["name"], "type": l.get("type", "")} for l in results.get("labels", [])]
+        print(json.dumps(labels, indent=2))
+        return
+
     service = build_service("gmail", "v1")
     results = service.users().labels().list(userId="me").execute()
     labels = [{"id": l["id"], "name": l["name"], "type": l.get("type", "")} for l in results.get("labels", [])]
     print(json.dumps(labels, indent=2))
 
 
+
 def gmail_modify(args):
-    service = build_service("gmail", "v1")
     body = {}
     if args.add_labels:
         body["addLabelIds"] = args.add_labels.split(",")
     if args.remove_labels:
         body["removeLabelIds"] = args.remove_labels.split(",")
+
+    if _gws_binary():
+        result = _run_gws(
+            ["gmail", "users", "messages", "modify"],
+            params={"userId": "me", "id": args.message_id},
+            body=body,
+        )
+        print(json.dumps({"id": result["id"], "labels": result.get("labelIds", [])}, indent=2))
+        return
+
+    service = build_service("gmail", "v1")
     result = service.users().messages().modify(userId="me", id=args.message_id, body=body).execute()
     print(json.dumps({"id": result["id"], "labels": result.get("labelIds", [])}, indent=2))
 
@@ -206,17 +418,40 @@ def gmail_modify(args):
 # Calendar
 # =========================================================================
 
+
 def calendar_list(args):
-    service = build_service("calendar", "v3")
     now = datetime.now(timezone.utc)
-    time_min = args.start or now.isoformat()
-    time_max = args.end or (now + timedelta(days=7)).isoformat()
+    time_min = _datetime_with_timezone(args.start or now.isoformat())
+    time_max = _datetime_with_timezone(args.end or (now + timedelta(days=7)).isoformat())
 
-    # Ensure timezone info
-    for val in [time_min, time_max]:
-        if "T" in val and "Z" not in val and "+" not in val and "-" not in val[11:]:
-            val += "Z"
+    if _gws_binary():
+        results = _run_gws(
+            ["calendar", "events", "list"],
+            params={
+                "calendarId": args.calendar,
+                "timeMin": time_min,
+                "timeMax": time_max,
+                "maxResults": args.max,
+                "singleEvents": True,
+                "orderBy": "startTime",
+            },
+        )
+        events = []
+        for e in results.get("items", []):
+            events.append({
+                "id": e["id"],
+                "summary": e.get("summary", "(no title)"),
+                "start": e.get("start", {}).get("dateTime", e.get("start", {}).get("date", "")),
+                "end": e.get("end", {}).get("dateTime", e.get("end", {}).get("date", "")),
+                "location": e.get("location", ""),
+                "description": e.get("description", ""),
+                "status": e.get("status", ""),
+                "htmlLink": e.get("htmlLink", ""),
+            })
+        print(json.dumps(events, indent=2, ensure_ascii=False))
+        return
 
+    service = build_service("calendar", "v3")
     results = service.events().list(
         calendarId=args.calendar, timeMin=time_min, timeMax=time_max,
         maxResults=args.max, singleEvents=True, orderBy="startTime",
@@ -237,8 +472,8 @@ def calendar_list(args):
     print(json.dumps(events, indent=2, ensure_ascii=False))
 
 
+
 def calendar_create(args):
-    service = build_service("calendar", "v3")
     event = {
         "summary": args.summary,
         "start": {"dateTime": args.start},
@@ -249,8 +484,23 @@ def calendar_create(args):
     if args.description:
         event["description"] = args.description
     if args.attendees:
-        event["attendees"] = [{"email": e.strip()} for e in args.attendees.split(",")]
+        event["attendees"] = [{"email": e.strip()} for e in args.attendees.split(",") if e.strip()]
 
+    if _gws_binary():
+        result = _run_gws(
+            ["calendar", "events", "insert"],
+            params={"calendarId": args.calendar},
+            body=event,
+        )
+        print(json.dumps({
+            "status": "created",
+            "id": result["id"],
+            "summary": result.get("summary", ""),
+            "htmlLink": result.get("htmlLink", ""),
+        }, indent=2))
+        return
+
+    service = build_service("calendar", "v3")
     result = service.events().insert(calendarId=args.calendar, body=event).execute()
     print(json.dumps({
         "status": "created",
@@ -260,7 +510,13 @@ def calendar_create(args):
     }, indent=2))
 
 
+
 def calendar_delete(args):
+    if _gws_binary():
+        _run_gws(["calendar", "events", "delete"], params={"calendarId": args.calendar, "eventId": args.event_id})
+        print(json.dumps({"status": "deleted", "eventId": args.event_id}))
+        return
+
     service = build_service("calendar", "v3")
     service.events().delete(calendarId=args.calendar, eventId=args.event_id).execute()
     print(json.dumps({"status": "deleted", "eventId": args.event_id}))
@@ -270,9 +526,22 @@ def calendar_delete(args):
 # Drive
 # =========================================================================
 
+
 def drive_search(args):
+    query = args.query if args.raw_query else f"fullText contains '{args.query}'"
+    if _gws_binary():
+        results = _run_gws(
+            ["drive", "files", "list"],
+            params={
+                "q": query,
+                "pageSize": args.max,
+                "fields": "files(id, name, mimeType, modifiedTime, webViewLink)",
+            },
+        )
+        print(json.dumps(results.get("files", []), indent=2, ensure_ascii=False))
+        return
+
     service = build_service("drive", "v3")
-    query = f"fullText contains '{args.query}'" if not args.raw_query else args.query
     results = service.files().list(
         q=query, pageSize=args.max, fields="files(id, name, mimeType, modifiedTime, webViewLink)",
     ).execute()
@@ -284,7 +553,30 @@ def drive_search(args):
 # Contacts
 # =========================================================================
 
+
 def contacts_list(args):
+    if _gws_binary():
+        results = _run_gws(
+            ["people", "people", "connections", "list"],
+            params={
+                "resourceName": "people/me",
+                "pageSize": args.max,
+                "personFields": "names,emailAddresses,phoneNumbers",
+            },
+        )
+        contacts = []
+        for person in results.get("connections", []):
+            names = person.get("names", [{}])
+            emails = person.get("emailAddresses", [])
+            phones = person.get("phoneNumbers", [])
+            contacts.append({
+                "name": names[0].get("displayName", "") if names else "",
+                "emails": [e.get("value", "") for e in emails],
+                "phones": [p.get("value", "") for p in phones],
+            })
+        print(json.dumps(contacts, indent=2, ensure_ascii=False))
+        return
+
     service = build_service("people", "v1")
     results = service.people().connections().list(
         resourceName="people/me",
@@ -308,7 +600,16 @@ def contacts_list(args):
 # Sheets
 # =========================================================================
 
+
 def sheets_get(args):
+    if _gws_binary():
+        result = _run_gws(
+            ["sheets", "spreadsheets", "values", "get"],
+            params={"spreadsheetId": args.sheet_id, "range": args.range},
+        )
+        print(json.dumps(result.get("values", []), indent=2, ensure_ascii=False))
+        return
+
     service = build_service("sheets", "v4")
     result = service.spreadsheets().values().get(
         spreadsheetId=args.sheet_id, range=args.range,
@@ -316,10 +617,25 @@ def sheets_get(args):
     print(json.dumps(result.get("values", []), indent=2, ensure_ascii=False))
 
 
+
 def sheets_update(args):
-    service = build_service("sheets", "v4")
     values = json.loads(args.values)
     body = {"values": values}
+
+    if _gws_binary():
+        result = _run_gws(
+            ["sheets", "spreadsheets", "values", "update"],
+            params={
+                "spreadsheetId": args.sheet_id,
+                "range": args.range,
+                "valueInputOption": "USER_ENTERED",
+            },
+            body=body,
+        )
+        print(json.dumps({"updatedCells": result.get("updatedCells", 0), "updatedRange": result.get("updatedRange", "")}, indent=2))
+        return
+
+    service = build_service("sheets", "v4")
     result = service.spreadsheets().values().update(
         spreadsheetId=args.sheet_id, range=args.range,
         valueInputOption="USER_ENTERED", body=body,
@@ -327,10 +643,26 @@ def sheets_update(args):
     print(json.dumps({"updatedCells": result.get("updatedCells", 0), "updatedRange": result.get("updatedRange", "")}, indent=2))
 
 
+
 def sheets_append(args):
-    service = build_service("sheets", "v4")
     values = json.loads(args.values)
     body = {"values": values}
+
+    if _gws_binary():
+        result = _run_gws(
+            ["sheets", "spreadsheets", "values", "append"],
+            params={
+                "spreadsheetId": args.sheet_id,
+                "range": args.range,
+                "valueInputOption": "USER_ENTERED",
+                "insertDataOption": "INSERT_ROWS",
+            },
+            body=body,
+        )
+        print(json.dumps({"updatedCells": result.get("updates", {}).get("updatedCells", 0)}, indent=2))
+        return
+
+    service = build_service("sheets", "v4")
     result = service.spreadsheets().values().append(
         spreadsheetId=args.sheet_id, range=args.range,
         valueInputOption="USER_ENTERED", insertDataOption="INSERT_ROWS", body=body,
@@ -342,21 +674,24 @@ def sheets_append(args):
 # Docs
 # =========================================================================
 
+
 def docs_get(args):
+    if _gws_binary():
+        doc = _run_gws(["docs", "documents", "get"], params={"documentId": args.doc_id})
+        result = {
+            "title": doc.get("title", ""),
+            "documentId": doc.get("documentId", ""),
+            "body": _extract_doc_text(doc),
+        }
+        print(json.dumps(result, indent=2, ensure_ascii=False))
+        return
+
     service = build_service("docs", "v1")
     doc = service.documents().get(documentId=args.doc_id).execute()
-    # Extract plain text from the document structure
-    text_parts = []
-    for element in doc.get("body", {}).get("content", []):
-        paragraph = element.get("paragraph", {})
-        for pe in paragraph.get("elements", []):
-            text_run = pe.get("textRun", {})
-            if text_run.get("content"):
-                text_parts.append(text_run["content"])
     result = {
         "title": doc.get("title", ""),
         "documentId": doc.get("documentId", ""),
-        "body": "".join(text_parts),
+        "body": _extract_doc_text(doc),
     }
     print(json.dumps(result, indent=2, ensure_ascii=False))
 
@@ -365,6 +700,7 @@ def docs_get(args):
 # CLI parser
 # =========================================================================
 
+
 def main():
     parser = argparse.ArgumentParser(description="Google Workspace API for Hermes Agent")
     sub = parser.add_subparsers(dest="service", required=True)
diff --git a/tests/skills/test_google_api_cli.py b/tests/skills/test_google_api_cli.py
new file mode 100644
index 00000000000..97b2e839557
--- /dev/null
+++ b/tests/skills/test_google_api_cli.py
@@ -0,0 +1,202 @@
+"""Tests for the Google Workspace skill CLI wrapper.
+
+These focus on the hybrid backend: prefer the Google Workspace CLI (`gws`) when
+available, while preserving the existing Hermes-facing JSON contract.
+"""
+
+import importlib.util
+import json
+from argparse import Namespace
+from pathlib import Path
+from types import SimpleNamespace
+
+import pytest
+
+
+SCRIPT_PATH = (
+    Path(__file__).resolve().parents[2]
+    / "skills/productivity/google-workspace/scripts/google_api.py"
+)
+
+
+def _load_module():
+    spec = importlib.util.spec_from_file_location("google_workspace_api_test", SCRIPT_PATH)
+    module = importlib.util.module_from_spec(spec)
+    assert spec.loader is not None
+    spec.loader.exec_module(module)
+    return module
+
+
+def _completed(stdout: str = "", stderr: str = "", returncode: int = 0):
+    return SimpleNamespace(stdout=stdout, stderr=stderr, returncode=returncode)
+
+
+@pytest.fixture
+def google_api_module(tmp_path, monkeypatch):
+    module = _load_module()
+    token_path = tmp_path / "google_token.json"
+    token_path.write_text(
+        json.dumps(
+            {
+                "token": "access-token",
+                "refresh_token": "refresh-token",
+                "token_uri": "https://oauth2.googleapis.com/token",
+                "client_id": "client-id",
+                "client_secret": "client-secret",
+            }
+        )
+    )
+    monkeypatch.setattr(module, "TOKEN_PATH", token_path)
+    monkeypatch.setattr(module, "_gws_binary", lambda: "/usr/bin/gws", raising=False)
+    monkeypatch.setattr(
+        module,
+        "build_service",
+        lambda *args, **kwargs: (_ for _ in ()).throw(AssertionError("legacy backend should not be used")),
+    )
+    return module
+
+
+def test_gmail_search_uses_gws_and_normalizes_results(google_api_module, monkeypatch, capsys):
+    calls = []
+
+    def fake_run(cmd, capture_output, text, env):
+        calls.append({"cmd": cmd, "env": env})
+        if cmd[1:4] == ["gmail", "users", "messages"] and cmd[4] == "list":
+            assert json.loads(cmd[6]) == {"userId": "me", "q": "is:unread", "maxResults": 5}
+            return _completed(
+                json.dumps({"messages": [{"id": "msg-1", "threadId": "thread-1"}]})
+            )
+        if cmd[1:4] == ["gmail", "users", "messages"] and cmd[4] == "get":
+            params = json.loads(cmd[6])
+            assert params["id"] == "msg-1"
+            return _completed(
+                json.dumps(
+                    {
+                        "id": "msg-1",
+                        "threadId": "thread-1",
+                        "payload": {
+                            "headers": [
+                                {"name": "From", "value": "alice@example.com"},
+                                {"name": "To", "value": "bob@example.com"},
+                                {"name": "Subject", "value": "Hello"},
+                                {"name": "Date", "value": "Sat, 15 Mar 2026 10:00:00 +0000"},
+                            ]
+                        },
+                        "snippet": "preview",
+                        "labelIds": ["UNREAD"],
+                    }
+                )
+            )
+        raise AssertionError(f"Unexpected command: {cmd}")
+
+    monkeypatch.setattr(google_api_module.subprocess, "run", fake_run)
+
+    google_api_module.gmail_search(Namespace(query="is:unread", max=5))
+
+    out = json.loads(capsys.readouterr().out)
+    assert out == [
+        {
+            "id": "msg-1",
+            "threadId": "thread-1",
+            "from": "alice@example.com",
+            "to": "bob@example.com",
+            "subject": "Hello",
+            "date": "Sat, 15 Mar 2026 10:00:00 +0000",
+            "snippet": "preview",
+            "labels": ["UNREAD"],
+        }
+    ]
+    assert calls[0]["env"]["GOOGLE_WORKSPACE_CLI_CREDENTIALS_FILE"] == str(google_api_module.TOKEN_PATH)
+
+
+def test_calendar_create_uses_gws_insert_and_normalizes_result(google_api_module, monkeypatch, capsys):
+    def fake_run(cmd, capture_output, text, env):
+        assert cmd[:5] == ["/usr/bin/gws", "calendar", "events", "insert", "--params"]
+        assert json.loads(cmd[5]) == {"calendarId": "primary"}
+        body = json.loads(cmd[7])
+        assert body == {
+            "summary": "Standup",
+            "start": {"dateTime": "2026-03-15T09:00:00Z"},
+            "end": {"dateTime": "2026-03-15T09:30:00Z"},
+            "location": "Room 1",
+            "description": "Daily sync",
+            "attendees": [{"email": "alice@example.com"}, {"email": "bob@example.com"}],
+        }
+        return _completed(json.dumps({"id": "evt-1", "summary": "Standup", "htmlLink": "https://calendar/event"}))
+
+    monkeypatch.setattr(google_api_module.subprocess, "run", fake_run)
+
+    google_api_module.calendar_create(
+        Namespace(
+            summary="Standup",
+            start="2026-03-15T09:00:00Z",
+            end="2026-03-15T09:30:00Z",
+            location="Room 1",
+            description="Daily sync",
+            attendees="alice@example.com,bob@example.com",
+            calendar="primary",
+        )
+    )
+
+    assert json.loads(capsys.readouterr().out) == {
+        "status": "created",
+        "id": "evt-1",
+        "summary": "Standup",
+        "htmlLink": "https://calendar/event",
+    }
+
+
+def test_sheets_append_uses_gws_and_returns_updated_cells(google_api_module, monkeypatch, capsys):
+    def fake_run(cmd, capture_output, text, env):
+        assert cmd[:6] == ["/usr/bin/gws", "sheets", "spreadsheets", "values", "append", "--params"]
+        assert json.loads(cmd[6]) == {
+            "spreadsheetId": "sheet-123",
+            "range": "Sheet1!A:C",
+            "valueInputOption": "USER_ENTERED",
+            "insertDataOption": "INSERT_ROWS",
+        }
+        assert json.loads(cmd[8]) == {"values": [["a", "b", "c"]]}
+        return _completed(json.dumps({"updates": {"updatedCells": 3}}))
+
+    monkeypatch.setattr(google_api_module.subprocess, "run", fake_run)
+
+    google_api_module.sheets_append(
+        Namespace(sheet_id="sheet-123", range="Sheet1!A:C", values='[["a", "b", "c"]]')
+    )
+
+    assert json.loads(capsys.readouterr().out) == {"updatedCells": 3}
+
+
+def test_docs_get_uses_gws_and_extracts_plain_text(google_api_module, monkeypatch, capsys):
+    def fake_run(cmd, capture_output, text, env):
+        assert cmd[:6] == ["/usr/bin/gws", "docs", "documents", "get", "--params", '{"documentId": "doc-123"}']
+        return _completed(
+            json.dumps(
+                {
+                    "title": "Doc Title",
+                    "documentId": "doc-123",
+                    "body": {
+                        "content": [
+                            {
+                                "paragraph": {
+                                    "elements": [
+                                        {"textRun": {"content": "Hello "}},
+                                        {"textRun": {"content": "world"}},
+                                    ]
+                                }
+                            }
+                        ]
+                    },
+                }
+            )
+        )
+
+    monkeypatch.setattr(google_api_module.subprocess, "run", fake_run)
+
+    google_api_module.docs_get(Namespace(doc_id="doc-123"))
+
+    assert json.loads(capsys.readouterr().out) == {
+        "title": "Doc Title",
+        "documentId": "doc-123",
+        "body": "Hello world",
+    }
diff --git a/website/docs/reference/skills-catalog.md b/website/docs/reference/skills-catalog.md
index 7e128f11fba..68bc4d98b8f 100644
--- a/website/docs/reference/skills-catalog.md
+++ b/website/docs/reference/skills-catalog.md
@@ -204,7 +204,7 @@ Skills for document creation, presentations, spreadsheets, and other productivit
 
 | Skill | Description | Path |
 |-------|-------------|------|
-| `google-workspace` | Gmail, Calendar, Drive, Contacts, Sheets, and Docs integration via Python. Uses OAuth2 with automatic token refresh. No external binaries needed — runs entirely with Google's Python client libraries in the Hermes venv. | `productivity/google-workspace` |
+| `google-workspace` | Gmail, Calendar, Drive, Contacts, Sheets, and Docs integration for Hermes. Uses Hermes-managed OAuth2 setup, prefers the Google Workspace CLI (`gws`) when available for broader API coverage, and falls back to the Python client libraries otherwise. | `productivity/google-workspace` |
 | `nano-pdf` | Edit PDFs with natural-language instructions using the nano-pdf CLI. Modify text, fix typos, update titles, and make content changes to specific pages without manual editing. | `productivity/nano-pdf` |
 | `notion` | Notion API for creating and managing pages, databases, and blocks via curl. Search, create, update, and query Notion workspaces directly from the terminal. | `productivity/notion` |
 | `ocr-and-documents` | Extract text from PDFs and scanned documents. Use web_extract for remote URLs, pymupdf for local text-based PDFs, marker-pdf for OCR/scanned docs. For DOCX use python-docx, for PPTX see the powerpoint skill. | `productivity/ocr-and-documents` |