Upload SpecStream tool

README.md

@@ -1,10 +1,12 @@
 ---
-title: Specstream Web Search
-emoji: 💻
-colorFrom: green
-colorTo: red
+title: SpecStream web search
 sdk: static
 pinned: false
+tags:
+- smolagents
+- tool
 ---
 
-Check out the configuration reference at https://huggingface.co/docs/hub/spaces-config-reference
+SpecStream is the live web data engine for teams and agents. One platform to build, search, extract, enrich, map, and monitor data across the entire web. Four tools: `specstream_web_search` (`GET /v1/web-search`, $5 per 1,000 searches), `specstream_extract_pages` (`POST /v1/extract`, $0.80 per 1,000 pages), `specstream_find_web_data` (Deep Search, priced per row) and `specstream_get_rows`. Source URL and read time on every value. Set SPECSTREAM_API_KEY. Start at https://specstream.ai/?utm_source=smolagents
+
+Load it with `Tool.from_hub("SpecStreamAI/specstream-web-search", trust_remote_code=True)`.

index.html

@@ -1,19 +1 @@
-<!doctype html>
-<html>
-	<head>
-		<meta charset="utf-8" />
-		<meta name="viewport" content="width=device-width" />
-		<title>My static Space</title>
-		<link rel="stylesheet" href="style.css" />
-	</head>
-	<body>
-		<div class="card">
-			<h1>Welcome to your static Space!</h1>
-			<p>You can modify this app directly by editing <i>index.html</i> in the Files and versions tab.</p>
-			<p>
-				Also don't forget to check the
-				<a href="https://huggingface.co/docs/hub/spaces" target="_blank">Spaces documentation</a>.
-			</p>
-		</div>
-	</body>
-</html>
+<!doctype html><html><head><meta charset="utf-8"><title>SpecStream web search</title></head><body style="font-family:system-ui;max-width:40rem;margin:3rem auto;padding:0 1rem"><h1>SpecStream web search</h1><p>SpecStream is the live web data engine for teams and agents. One platform to build, search, extract, enrich, map, and monitor data across the entire web. Four tools: `specstream_web_search` (`GET /v1/web-search`, $5 per 1,000 searches), `specstream_extract_pages` (`POST /v1/extract`, $0.80 per 1,000 pages), `specstream_find_web_data` (Deep Search, priced per row) and `specstream_get_rows`. Source URL and read time on every value. Set SPECSTREAM_API_KEY. Start at https://specstream.ai/?utm_source=smolagents</p><p><code>Tool.from_hub("SpecStreamAI/specstream-web-search", trust_remote_code=True)</code></p></body></html>

requirements.txt

@@ -0,0 +1 @@
+smolagents

tool.py

@@ -0,0 +1,50 @@
+from typing import Any, Optional
+from smolagents.tools import Tool
+import os
+import urllib
+import json
+
+class SpecStreamWebSearch(Tool):
+    name = "specstream_web_search"
+    description = "Searches the live web in seconds with the SpecStream Search API (GET /v1/web-search, $5 per 1,000 searches, no quote needed). Returns the answer with its source URL and read time, matching rows, ranked sources and optional page passages. Needs SPECSTREAM_API_KEY."
+    inputs = {'q': {'type': 'string', 'description': 'The search question or request in everyday words (1 to 2,000 characters).'}, 'passages': {'type': 'boolean', 'description': 'Set True to include the main text passages of the top matching pages.', 'nullable': True}}
+    output_type = "string"
+
+    def forward(self, q: str, passages: bool | None = None) -> str:
+        import json
+        import os
+        from urllib.error import HTTPError, URLError
+        from urllib.parse import urlencode
+        from urllib.request import Request, urlopen
+
+        key = os.environ.get("SPECSTREAM_API_KEY", "").strip()
+        if not key:
+            return "Set SPECSTREAM_API_KEY to a SpecStream API key (https://specstream.ai/workspace?utm_source=smolagents)."
+        query = (q or "").strip()
+        if not query or len(query) > 2000:
+            return "Provide a search query between 1 and 2,000 characters."
+        base = os.environ.get("SPECSTREAM_API_BASE", "https://api.specstream.ai").rstrip("/")
+        params = {"q": query}
+        if passages:
+            params["passages"] = "true"
+        req = Request(
+            base + "/v1/web-search?" + urlencode(params),
+            method="GET",
+            headers={
+                "Authorization": "Bearer " + key,
+                "Accept": "application/json",
+                "User-Agent": "specstream-smolagents/0.2",
+            },
+        )
+        try:
+            with urlopen(req, timeout=60) as resp:
+                payload = json.loads(resp.read().decode("utf-8"))
+        except HTTPError as exc:
+            return "SpecStream answered HTTP %d" % exc.code
+        except (URLError, OSError):
+            return "SpecStream could not be reached"
+        data = payload.get("data") if isinstance(payload, dict) else {}
+        return json.dumps(data, ensure_ascii=False)
+
+    def __init__(self, *args, **kwargs):
+        self.is_initialized = False