serpapi · adarshdigievo · Aug 4, 2026 · Aug 4, 2026 · Aug 4, 2026 · Aug 4, 2026
diff --git a/README.md b/README.md
@@ -26,7 +26,7 @@ When one supported agent SDK is installed, the package detects it and creates to
 
 ## Supported agent SDKs
 
-| SDK                       | Install extra / provider    | Returned tool                            |
+| SDK                       | Install extra               | Returned tool                            |
 |---------------------------|-----------------------------|------------------------------------------|
 | OpenAI Agents SDK         | `openai-agents`             | OpenAI Agents `FunctionTool`             |
 | Pydantic AI               | `pydantic-ai`               | Pydantic AI `Tool`                       |
@@ -98,7 +98,7 @@ result = Runner.run_sync(
 print(result.final_output)
 ```
 
-## Quickstart: explicit LangChain provider
+## Quickstart: LangChain
 
 Install the LangChain extra and the model backend used by this example:
 
@@ -120,9 +120,9 @@ from serpapi_search_tools import maps_search, news_search, web_search
 agent = create_agent(
     model=ChatOpenAI(model="gpt-5.4-mini", temperature=0),
     tools=[
-        web_search(provider="langchain"),
-        news_search(provider="langchain"),
-        maps_search(provider="langchain"),
+        web_search(),
+        news_search(),
+        maps_search(),
     ],
 )
 
@@ -142,10 +142,8 @@ result = agent.invoke(
 print(result["messages"][-1].content)
 ```
 
-When LangChain is the only supported SDK installed, these constructors are also auto-detected, so `web_search()` is
-enough. Supplying
-`provider="langchain"` explicitly is useful when several supported SDKs share an environment or when you want the
-integration choice to be visible in code.
+The constructors use automatic SDK detection, just as in the first quickstart. For multi-SDK environments and explicit
+selection, see [Agent SDKs](https://serpapi.github.io/serpapi-search-tools-python/user-guide/frameworks.html).
 
 For a step-by-step explanation, keys, customization, and troubleshooting, read
 the [detailed quickstart](https://serpapi.github.io/serpapi-search-tools-python/user-guide/quickstart.html).
@@ -196,9 +194,12 @@ tools = [
 ]
 ```
 
-`news_search` supports keyword searches in Google News. `maps_search` searches Google Maps and accepts optional
-`location`, `zoom` (`3` through `30`), and
-`nearby` fields. Place details, reviews, and directions use different SerpApi APIs and are not part of this search tool.
+`news_search` supports keyword searches in Google News. `maps_search` searches
+Google Maps and accepts optional `location`, `zoom` (`3` through `30`), and
+`nearby` fields. Use `nearby=True` for “near me” intent with a separate
+`location`; leave it false when the query already names a city or area. Place
+details, reviews, and directions use different SerpApi APIs and are not part of
+this search tool.
 
 ### Shopping
 
@@ -219,60 +220,46 @@ The package routes one human `query` to each marketplace's native field:
 Google Shopping uses `q`, Amazon uses `k`, Walmart uses `query`, and eBay uses
 `_nkw`.
 
-### Hotels
+### Travel
 
 ```python
-from serpapi_search_tools import hotels_search
-
-hotels = hotels_search(provider="function")
-result = hotels(
-    query="hotels in Kyoto",
-    check_in_date="2026-08-01",
-    check_out_date="2026-08-04",
-    adults=2,
-    children=1,
-    children_ages=[8],
-)
+from serpapi_search_tools import flights_search, hotels_search, travel_explore_search
+
+travel_tools = [
+    hotels_search(),
+    flights_search(),
+    travel_explore_search(),
+]
 ```
 
-Hotel dates use `YYYY-MM-DD`. Checkout must be after check-in. When `children`
-is nonzero, provide exactly one age from 1 through 17 per child.
+These constructors create hotel, flight, and destination-discovery tools for the detected agent SDK.
 
-### Flights
+#### Hotels
 
-```python
-from serpapi_search_tools import TravelClass, flights_search
-
-flights = flights_search(provider="function")
-result = flights(
-    departure_id="LAX",
-    arrival_id="AUS",
-    outbound_date="2026-08-01",
-    return_date="2026-08-04",
-    travel_class=TravelClass.BUSINESS,
-    adults=1,
-)
-```
-
-`flights_search` requires an origin, destination, and outbound date. Omitting
-`return_date` creates a one-way request; including it creates a round trip. Multi-city searches are not currently
-supported.
+`hotels_search` lets the agent provide a destination, dates, and guest details. Hotel dates use `YYYY-MM-DD`. Checkout
+must be after check-in. When `children`
+is nonzero, provide exactly one age from 1 through 17 per child; use `1` for a
+child under one year old.
 
-### Explore destinations
+#### Flights
 
-```python
-from serpapi_search_tools import travel_explore_search
+`flights_search` lets the agent provide route, date, cabin, and passenger details. It requires an origin, destination,
+and outbound date. Omitting
+`return_date` creates a one-way request; including it creates a round trip.
+Use specific airport IATA codes such as `LHR` and `CDG`, not metropolitan codes
+such as `LON` and `PAR`. For a city-wide search, use a Google Knowledge Graph
+location ID (KGMID) beginning with `/m/` or `/g/`, such as `/m/04jpl` for
+London or `/m/05qtj` for Paris. Multi-city searches are not currently
+supported.
 
-explore = travel_explore_search(provider="function")
-result = explore(
-    departure_id="JFK",
-    arrival_area_id="/m/02j9z",
-)
-```
+#### Explore destinations
 
-Travel Explore requires only a departure identifier. It can also accept an arrival identifier or area, fixed
-outbound/return dates, cabin class, and passenger counts. These travel tools send their route and date fields directly
-to the matching SerpApi endpoint.
+`travel_explore_search` lets the agent discover destinations and requires only a departure airport IATA code or city
+KGMID. It
+can also accept a specific arrival airport or city through `arrival_id`, or a
+region or country KGMID through `arrival_area_id`. Fixed outbound/return dates,
+cabin class, and passenger counts are optional. These travel tools send their
+route and date fields directly to the matching SerpApi endpoint.
 
 ## Set advanced parameters in application code
 
@@ -296,35 +283,24 @@ defaults. Reserved keys (`api_key`, `async`, `engine`, and `output`) are rejecte
 
 ## Handle search failures
 
-The built-in client raises `SerpApiSearchError` when SerpApi rejects or cannot complete a request:
-
-```python
-from serpapi_search_tools import SerpApiSearchError, web_search
-
-search = web_search(provider="function")
-try:
-    result = search(query="Python packaging")
-except SerpApiSearchError as exc:
-    print(f"Search failed: {exc}")
-```
-
-Local input errors such as invalid dates or incompatible parameters remain
-`ValueError`, so applications can distinguish validation from provider and transport failures.
+The search runtime raises `SerpApiSearchError` for SerpApi and transport failures. Invalid tool inputs raise
+`ValueError`. Agent SDKs surface or handle tool errors differently, so use your SDK's normal tool-error mechanism. See
+[Debugging](https://serpapi.github.io/serpapi-search-tools-python/user-guide/debugging.html) for detailed examples.
 
 ## Common factory options
 
 Every constructor accepts:
 
-| Option             | Purpose                                                                                                           |
-|--------------------|-------------------------------------------------------------------------------------------------------------------|
-| `provider`         | Defaults to `"auto"`; select an SDK explicitly in multi-SDK environments or use `"function"` for a plain callable |
-| `include_examples` | Include or omit a short example in the model description                                                          |
-| `api_key`          | Explicit SerpApi key                                                                                              |
-| `client`           | Custom object with `search(params)` for caching, interception, or tests                                           |
-| `default_params`   | Application-controlled SerpApi options                                                                            |
-| `timeout`          | Timeout passed to the SerpApi SDK client                                                                          |
-| `name`             | Tool name presented to the model                                                                                  |
-| `mode`             | Result detail level: `"compact"` (default) or `"full"`                                                            |
+| Option             | Purpose                                                                                               |
+|--------------------|-------------------------------------------------------------------------------------------------------|
+| `provider`         | Defaults to `"auto"`; select an SDK explicitly only when multiple supported SDKs share an environment |
+| `include_examples` | Include or omit a short example in the model description                                              |
+| `api_key`          | Explicit SerpApi key                                                                                  |
+| `client`           | Custom object with `search(params)` for caching, interception, or tests                               |
+| `default_params`   | Application-controlled SerpApi options                                                                |
+| `timeout`          | Timeout passed to the SerpApi SDK client                                                              |
+| `name`             | Tool name presented to the model                                                                      |
+| `mode`             | Result detail level: `"compact"` (default) or `"full"`                                                |
 
 `web_search` and `shopping_search` additionally accept `allowed_engines` and
 `default_engine`. The tool offers only the engine values you configured.
@@ -348,8 +324,9 @@ Every constructor accepts:
 - [Google Flights](https://serpapi.com/google-flights-api)
 - [Google Travel Explore](https://serpapi.com/google-travel-explore-api)
 
-SerpApi's complete documentation index is available in
-[`llms.txt`](https://serpapi.com/llms.txt).
+For AI coding agents that need broader SerpApi API context, use
+[SerpApi's agent-friendly documentation index (`llms.txt`)](https://serpapi.com/llms.txt). It links directly to Markdown
+API references, including APIs beyond those wrapped by this package.
 
 ## More guides
 

diff --git a/cookbook/pydantic-ai/main.py b/cookbook/pydantic-ai/main.py
@@ -6,14 +6,14 @@
 
 from dotenv import load_dotenv
 from pydantic_ai import Agent
-from pydantic_ai.models.openai import OpenAIChatModel
+from pydantic_ai.models.openai import OpenAIResponsesModel
 from pydantic_ai.providers.openai import OpenAIProvider
 
 from serpapi_search_tools import images_search, maps_search, web_search
 
 load_dotenv()
 
-MODEL = os.getenv("OPENAI_MODEL", "gpt-5.4-mini")
+MODEL = os.getenv("OPENAI_MODEL", "gpt-5.6-luna")
 PROMPT = os.getenv(
     "COOKBOOK_PROMPT",
     (
@@ -48,7 +48,7 @@ def _write_report(text: str) -> Path:
 def main() -> None:
     _require_serpapi_key()
     provider = OpenAIProvider(api_key=_require_env("OPENAI_API_KEY"))
-    model = OpenAIChatModel(MODEL, provider=provider)
+    model = OpenAIResponsesModel(MODEL, provider=provider)
     agent = Agent(
         model,
         instructions=(

diff --git a/docs/cookbook/agno.qmd b/docs/cookbook/agno.qmd
@@ -5,18 +5,93 @@ description: Combine category facts, launches, and live price signals.
 
 # Agno market research
 
-This agent analyzes a product category using durable facts, recent launches,
-and current shopping listings, then writes a market report.
+- **Agent pattern:** Bounded tool loop
+- **Search surfaces:** Web, news, and shopping
+- **Saved artifact:** `agno-market-research.md`
 
-SerpApi enhancement: web, news, and shopping tools replace the upstream search
-dependency while preserving the decision-focused market research goal.
+## What you'll build
+
+An Agno agent that assesses the US market for home espresso grinders under
+USD 800. It separates durable category and manufacturer facts from recent
+launches and time-sensitive marketplace signals, then writes a decision-ready
+market report.
+
+**Default brief:** Map the category, representative products, gaps, and risks.
+Treat live price and availability as signals—not permanent product facts—and
+include source URLs.
+
+## How the agent works
+
+1. **Frame the category.** Web search establishes segments, manufacturers, and
+   durable product facts.
+2. **Check current signals.** News finds launches while shopping search captures
+   live prices and availability.
+3. **Reconcile the market.** The agent compares the evidence, calls out gaps,
+   and stays within a ten-call tool budget.
+
+## Core agent setup
+
+This abridged excerpt keeps the Agno agent and tool boundary in view. The full
+script also handles model configuration, environment validation, and Markdown
+output.
+
+```python
+agent = Agent(
+    name="SerpApi market researcher",
+    model=model,
+    instructions=[
+        "Search before making market claims.",
+        "Use the most specific SerpApi tool for each question.",
+        "Separate live listing signals from durable category facts.",
+    ],
+    tools=[
+        web_search(provider="agno", allowed_engines=["google_light", "bing"]),
+        news_search(provider="agno"),
+        shopping_search(provider="agno"),
+    ],
+    markdown=True,
+    tool_call_limit=10,
+)
+response = agent.run(PROMPT)
+```
+
+[View the complete `main.py` →](https://github.com/serpapi/serpapi-search-tools-python/blob/main/cookbook/agno/main.py)
+
+## Run the recipe
+
+Set `SERPAPI_API_KEY` and `XAI_API_KEY` in the repository-root `.env`, then
+run:
 
 ```bash
 uv run --isolated --no-project --with 'serpapi-search-tools[agno]' --with python-dotenv cookbook/agno/main.py
 ```
 
-Read the
-[complete cookbook entry](https://github.com/serpapi/serpapi-search-tools-python/tree/main/cookbook/agno)
-or Agno's official
-[market research cookbook](https://github.com/agno-agi/agno/blob/main/cookbook/91_tools/parallel/market_research.py).
+Optional controls: `XAI_MODEL`, `XAI_BASE_URL`, `COOKBOOK_PROMPT`, and
+`COOKBOOK_OUTPUT_DIR`.
+
+**Portable environment:** `--isolated --no-project` runs the local recipe
+against the published `serpapi-search-tools[agno]` package instead of importing
+this checkout.
+
+## Inspect the result
+
+The recipe writes `cookbook-output/agno-market-research.md`. Expect category
+segments, representative products, current price signals, market gaps, risks,
+and direct source URLs.
+
+The same report is printed to the terminal. Set `COOKBOOK_OUTPUT_DIR` to move
+the saved artifact, or `COOKBOOK_PROMPT` to research another category.
+
+## SerpApi tools used
+
+`web_search` supplies category and manufacturer facts, `news_search` tracks
+recent launches, and `shopping_search` captures current prices and
+availability. Keeping these result types separate helps the agent label which
+signals are durable and which may change quickly.
+
+**Framework pattern:** The agent structure is inspired by Agno's parallel
+market research cookbook; this recipe applies it to three SerpApi evidence
+surfaces.
 
+- [Open the runnable recipe](https://github.com/serpapi/serpapi-search-tools-python/tree/main/cookbook/agno)
+- [View Agno's source example](https://github.com/agno-agi/agno/blob/main/cookbook/91_tools/parallel/market_research.py)