feat(community): add Brave Search web search tool (#3528)

* feat(community): add Brave Search web search tool

Add a community web_search provider backed by the official Brave Search
API (https://api.search.brave.com/res/v1/web/search). API key is read
from the tool config (inline api_key) or the BRAVE_SEARCH_API_KEY env
var. Output schema (title/url/content) matches existing search tools.
No new dependencies (uses the existing httpx). Also wires up the setup
wizard, doctor health check, config example, and EN/ZH docs.

* refactor(community): drop redundant [:count] slice in Brave search

The Brave API already caps results via the `count` request param, so
client-side slicing was redundant. Tests now simulate the API honoring
`count` instead of relying on the slice. Addresses PR review nit.

* style(tests): apply ruff format to test_doctor.py

Collapse multiline write_text calls onto single lines to satisfy the
CI ruff formatter (lint-backend was failing on format --check).
This commit is contained in:
Ryker_Feng
2026-06-13 22:47:35 +08:00
committed by GitHub
parent 8955b3222a
commit 6e839342a7
10 changed files with 582 additions and 7 deletions
+2 -1
View File
@@ -234,7 +234,7 @@ tools:
```
**Built-in Tools**:
- `web_search` - Search the web (DuckDuckGo, Tavily, Exa, InfoQuest, Firecrawl)
- `web_search` - Search the web (DuckDuckGo, Tavily, Brave, Exa, InfoQuest, Firecrawl)
- `web_fetch` - Fetch web pages (Jina AI, Exa, InfoQuest, Firecrawl)
- `ls` - List directory contents
- `read_file` - Read file contents
@@ -364,6 +364,7 @@ models:
- `MIMO_API_KEY` - Xiaomi MiMo API key
- `NOVITA_API_KEY` - Novita API key (OpenAI-compatible endpoint)
- `TAVILY_API_KEY` - Tavily search API key
- `BRAVE_SEARCH_API_KEY` - Brave Search API key
- `DEER_FLOW_PROJECT_ROOT` - Project root for relative runtime paths
- `DEER_FLOW_CONFIG_PATH` - Custom config file path
- `DEER_FLOW_EXTENSIONS_CONFIG_PATH` - Custom extensions config file path
@@ -0,0 +1,3 @@
from .tools import web_search_tool
__all__ = ["web_search_tool"]
@@ -0,0 +1,119 @@
"""
Web Search Tool - Search the web using the Brave Search API.
Brave Search provides web results from an independent search index via a
REST API. An API key is required. Sign up at https://brave.com/search/api/
to get one.
Unlike the DuckDuckGo ``backend: brave`` option (which scrapes results via the
DDGS aggregator), this provider calls the official Brave Search API directly,
giving structured results, authenticated quota, and a documented SLA.
"""
import json
import logging
import os
import httpx
from langchain.tools import tool
from deerflow.config import get_app_config
logger = logging.getLogger(__name__)
_BRAVE_ENDPOINT = "https://api.search.brave.com/res/v1/web/search"
_DEFAULT_MAX_RESULTS = 5
# Brave Search API caps the `count` parameter at 20 results per request.
_BRAVE_MAX_COUNT = 20
_api_key_warned = False
def _get_api_key() -> str | None:
config = get_app_config().get_tool_config("web_search")
if config is not None:
api_key = (config.model_extra or {}).get("api_key")
if isinstance(api_key, str) and api_key.strip():
return api_key
return os.getenv("BRAVE_SEARCH_API_KEY")
def _coerce_max_results(value: object, *, default: int = _DEFAULT_MAX_RESULTS) -> int:
try:
coerced = int(value)
except (TypeError, ValueError):
logger.warning(
"Invalid Brave Search max_results=%r; using default %s",
value,
default,
)
coerced = default
return max(1, min(coerced, _BRAVE_MAX_COUNT))
@tool("web_search", parse_docstring=True)
def web_search_tool(query: str, max_results: int = 5) -> str:
"""Search the web for information using Brave Search.
Args:
query: Search keywords describing what you want to find. Be specific for better results.
max_results: Maximum number of search results to return. Default is 5.
"""
global _api_key_warned
config = get_app_config().get_tool_config("web_search")
if config is not None and "max_results" in (config.model_extra or {}):
max_results = config.model_extra["max_results"]
count = _coerce_max_results(max_results)
api_key = _get_api_key()
if not api_key:
if not _api_key_warned:
_api_key_warned = True
logger.warning("Brave Search API key is not set. Set BRAVE_SEARCH_API_KEY in your environment or provide api_key in config.yaml. Sign up at https://brave.com/search/api/")
return json.dumps(
{"error": "BRAVE_SEARCH_API_KEY is not configured", "query": query},
ensure_ascii=False,
)
headers = {
"X-Subscription-Token": api_key,
"Accept": "application/json",
}
params = {"q": query, "count": count, "text_decorations": False}
try:
with httpx.Client(timeout=30) as client:
response = client.get(_BRAVE_ENDPOINT, headers=headers, params=params)
response.raise_for_status()
data = response.json()
except httpx.HTTPStatusError as e:
logger.error(f"Brave Search API returned HTTP {e.response.status_code}: {e.response.text}")
return json.dumps(
{"error": f"Brave Search API error: HTTP {e.response.status_code}", "query": query},
ensure_ascii=False,
)
except Exception as e:
logger.error(f"Brave search failed: {type(e).__name__}: {e}")
return json.dumps({"error": str(e), "query": query}, ensure_ascii=False)
web_results = (data.get("web") or {}).get("results", [])
if not web_results:
return json.dumps({"error": "No results found", "query": query}, ensure_ascii=False)
normalized_results = [
{
"title": r.get("title", ""),
"url": r.get("url", ""),
"content": r.get("description", ""),
}
for r in web_results
]
output = {
"query": query,
"total_results": len(normalized_results),
"results": normalized_results,
}
return json.dumps(output, indent=2, ensure_ascii=False)
+360
View File
@@ -0,0 +1,360 @@
"""Unit tests for the Brave Search community web search tool."""
import json
from unittest.mock import MagicMock, patch
import httpx
import pytest
@pytest.fixture(autouse=True)
def reset_api_key_warned():
"""Reset the module-level warning flag before each test."""
import deerflow.community.brave.tools as brave_mod
brave_mod._api_key_warned = False
yield
brave_mod._api_key_warned = False
@pytest.fixture
def mock_config_with_key():
with patch("deerflow.community.brave.tools.get_app_config") as mock:
tool_config = MagicMock()
tool_config.model_extra = {"api_key": "test-brave-key", "max_results": 5}
mock.return_value.get_tool_config.return_value = tool_config
yield mock
@pytest.fixture
def mock_config_no_key():
with patch("deerflow.community.brave.tools.get_app_config") as mock:
tool_config = MagicMock()
tool_config.model_extra = {}
mock.return_value.get_tool_config.return_value = tool_config
yield mock
def _make_brave_response(results: list) -> MagicMock:
mock_resp = MagicMock()
mock_resp.json.return_value = {"web": {"results": results}}
mock_resp.raise_for_status = MagicMock()
return mock_resp
def _count_aware_get(results: list):
"""Mimic Brave returning at most `count` results for the request."""
def _get(url, **kwargs):
count = kwargs["params"]["count"]
return _make_brave_response(results[:count])
return _get
class TestGetApiKey:
def test_returns_config_key_when_present(self):
with patch("deerflow.community.brave.tools.get_app_config") as mock:
tool_config = MagicMock()
tool_config.model_extra = {"api_key": "from-config"}
mock.return_value.get_tool_config.return_value = tool_config
from deerflow.community.brave.tools import _get_api_key
assert _get_api_key() == "from-config"
def test_falls_back_to_env_when_config_key_empty(self):
with patch("deerflow.community.brave.tools.get_app_config") as mock:
tool_config = MagicMock()
tool_config.model_extra = {"api_key": " "}
mock.return_value.get_tool_config.return_value = tool_config
with patch.dict("os.environ", {"BRAVE_SEARCH_API_KEY": "env-key"}, clear=True):
from deerflow.community.brave.tools import _get_api_key
assert _get_api_key() == "env-key"
def test_falls_back_to_env_when_no_config(self):
with patch("deerflow.community.brave.tools.get_app_config") as mock:
mock.return_value.get_tool_config.return_value = None
with patch.dict("os.environ", {"BRAVE_SEARCH_API_KEY": "env-only"}, clear=True):
from deerflow.community.brave.tools import _get_api_key
assert _get_api_key() == "env-only"
def test_ignores_legacy_brave_api_key(self):
with patch("deerflow.community.brave.tools.get_app_config") as mock:
mock.return_value.get_tool_config.return_value = None
with patch.dict("os.environ", {"BRAVE_API_KEY": "legacy"}, clear=True):
from deerflow.community.brave.tools import _get_api_key
assert _get_api_key() is None
def test_returns_none_when_no_key_anywhere(self):
with patch("deerflow.community.brave.tools.get_app_config") as mock:
mock.return_value.get_tool_config.return_value = None
with patch.dict("os.environ", {}, clear=True):
from deerflow.community.brave.tools import _get_api_key
assert _get_api_key() is None
def test_model_extra_none_does_not_crash(self):
with patch("deerflow.community.brave.tools.get_app_config") as mock:
tool_config = MagicMock()
tool_config.model_extra = None
mock.return_value.get_tool_config.return_value = tool_config
with patch.dict("os.environ", {"BRAVE_SEARCH_API_KEY": "env-key"}, clear=True):
from deerflow.community.brave.tools import _get_api_key
assert _get_api_key() == "env-key"
class TestWebSearchTool:
def test_basic_search_returns_normalized_results(self, mock_config_with_key):
results = [
{"title": "Result 1", "url": "https://example.com/1", "description": "Desc 1"},
{"title": "Result 2", "url": "https://example.com/2", "description": "Desc 2"},
]
mock_resp = _make_brave_response(results)
with patch("deerflow.community.brave.tools.httpx.Client") as mock_client_cls:
mock_client_cls.return_value.__enter__.return_value.get.return_value = mock_resp
from deerflow.community.brave.tools import web_search_tool
result = web_search_tool.invoke({"query": "python tutorial"})
parsed = json.loads(result)
assert parsed["query"] == "python tutorial"
assert parsed["total_results"] == 2
assert parsed["results"][0]["title"] == "Result 1"
assert parsed["results"][0]["url"] == "https://example.com/1"
assert parsed["results"][0]["content"] == "Desc 1"
def test_respects_max_results_from_config(self, mock_config_with_key):
mock_config_with_key.return_value.get_tool_config.return_value.model_extra = {
"api_key": "test-key",
"max_results": 3,
}
results = [{"title": f"R{i}", "url": f"https://x.com/{i}", "description": f"D{i}"} for i in range(10)]
with patch("deerflow.community.brave.tools.httpx.Client") as mock_client_cls:
mock_client_cls.return_value.__enter__.return_value.get.side_effect = _count_aware_get(results)
from deerflow.community.brave.tools import web_search_tool
result = web_search_tool.invoke({"query": "test"})
parsed = json.loads(result)
assert parsed["total_results"] == 3
assert len(parsed["results"]) == 3
def test_max_results_parameter_accepted(self, mock_config_no_key):
"""Tool accepts max_results as a call parameter when config does not override it."""
results = [{"title": f"R{i}", "url": f"https://x.com/{i}", "description": f"D{i}"} for i in range(10)]
with patch.dict("os.environ", {"BRAVE_SEARCH_API_KEY": "env-key"}, clear=True):
with patch("deerflow.community.brave.tools.httpx.Client") as mock_client_cls:
mock_client_cls.return_value.__enter__.return_value.get.side_effect = _count_aware_get(results)
from deerflow.community.brave.tools import web_search_tool
result = web_search_tool.invoke({"query": "test", "max_results": 2})
parsed = json.loads(result)
assert parsed["total_results"] == 2
def test_config_max_results_overrides_parameter(self):
with patch("deerflow.community.brave.tools.get_app_config") as mock:
tool_config = MagicMock()
tool_config.model_extra = {"api_key": "test-key", "max_results": 3}
mock.return_value.get_tool_config.return_value = tool_config
results = [{"title": f"R{i}", "url": f"https://x.com/{i}", "description": f"D{i}"} for i in range(10)]
with patch("deerflow.community.brave.tools.httpx.Client") as mock_client_cls:
mock_client_cls.return_value.__enter__.return_value.get.side_effect = _count_aware_get(results)
from deerflow.community.brave.tools import web_search_tool
result = web_search_tool.invoke({"query": "test", "max_results": 8})
parsed = json.loads(result)
assert parsed["total_results"] == 3
def test_max_results_string_from_env_is_coerced_and_clamped(self):
"""Env-sourced max_results is a string and must be coerced and clamped to 20."""
with patch("deerflow.community.brave.tools.get_app_config") as mock:
tool_config = MagicMock()
tool_config.model_extra = {"api_key": "test-key", "max_results": "50"}
mock.return_value.get_tool_config.return_value = tool_config
results = [{"title": f"R{i}", "url": f"https://x.com/{i}", "description": f"D{i}"} for i in range(30)]
with patch("deerflow.community.brave.tools.httpx.Client") as mock_client_cls:
mock_get = mock_client_cls.return_value.__enter__.return_value.get
mock_get.side_effect = _count_aware_get(results)
from deerflow.community.brave.tools import web_search_tool
result = web_search_tool.invoke({"query": "test"})
parsed = json.loads(result)
params = mock_get.call_args.kwargs["params"]
assert params["count"] == 20
assert parsed["total_results"] == 20
def test_invalid_max_results_falls_back_to_default(self, caplog):
with patch("deerflow.community.brave.tools.get_app_config") as mock:
tool_config = MagicMock()
tool_config.model_extra = {"api_key": "test-key", "max_results": "abc"}
mock.return_value.get_tool_config.return_value = tool_config
results = [{"title": f"R{i}", "url": f"https://x.com/{i}", "description": f"D{i}"} for i in range(10)]
with patch("deerflow.community.brave.tools.httpx.Client") as mock_client_cls:
mock_get = mock_client_cls.return_value.__enter__.return_value.get
mock_get.side_effect = _count_aware_get(results)
from deerflow.community.brave.tools import web_search_tool
with caplog.at_level("WARNING", logger="deerflow.community.brave.tools"):
result = web_search_tool.invoke({"query": "test"})
parsed = json.loads(result)
params = mock_get.call_args.kwargs["params"]
assert params["count"] == 5
assert parsed["total_results"] == 5
assert any("Invalid Brave Search max_results" in record.message for record in caplog.records)
def test_empty_results_returns_error_json(self, mock_config_with_key):
mock_resp = _make_brave_response([])
with patch("deerflow.community.brave.tools.httpx.Client") as mock_client_cls:
mock_client_cls.return_value.__enter__.return_value.get.return_value = mock_resp
from deerflow.community.brave.tools import web_search_tool
result = web_search_tool.invoke({"query": "no results"})
parsed = json.loads(result)
assert parsed["error"] == "No results found"
assert parsed["query"] == "no results"
def test_missing_web_key_returns_error_json(self, mock_config_with_key):
"""A response without a `web` block should be treated as no results."""
mock_resp = MagicMock()
mock_resp.json.return_value = {}
mock_resp.raise_for_status = MagicMock()
with patch("deerflow.community.brave.tools.httpx.Client") as mock_client_cls:
mock_client_cls.return_value.__enter__.return_value.get.return_value = mock_resp
from deerflow.community.brave.tools import web_search_tool
result = web_search_tool.invoke({"query": "test"})
parsed = json.loads(result)
assert parsed["error"] == "No results found"
def test_missing_api_key_returns_error_json(self, mock_config_no_key):
with patch.dict("os.environ", {}, clear=True):
from deerflow.community.brave.tools import web_search_tool
result = web_search_tool.invoke({"query": "test"})
parsed = json.loads(result)
assert "error" in parsed
assert "BRAVE_SEARCH_API_KEY" in parsed["error"]
def test_missing_api_key_logs_warning_once(self, mock_config_no_key, caplog):
import logging
with patch.dict("os.environ", {}, clear=True):
from deerflow.community.brave.tools import web_search_tool
with caplog.at_level(logging.WARNING, logger="deerflow.community.brave.tools"):
web_search_tool.invoke({"query": "q1"})
web_search_tool.invoke({"query": "q2"})
warnings = [r for r in caplog.records if r.levelno == logging.WARNING]
assert len(warnings) == 1
def test_http_error_returns_structured_error(self, mock_config_with_key):
mock_error_response = MagicMock()
mock_error_response.status_code = 403
mock_error_response.text = "Forbidden"
with patch("deerflow.community.brave.tools.httpx.Client") as mock_client_cls:
mock_client_cls.return_value.__enter__.return_value.get.side_effect = httpx.HTTPStatusError("403", request=MagicMock(), response=mock_error_response)
from deerflow.community.brave.tools import web_search_tool
result = web_search_tool.invoke({"query": "test"})
parsed = json.loads(result)
assert "error" in parsed
assert "403" in parsed["error"]
def test_network_exception_returns_error_json(self, mock_config_with_key):
with patch("deerflow.community.brave.tools.httpx.Client") as mock_client_cls:
mock_client_cls.return_value.__enter__.return_value.get.side_effect = Exception("timeout")
from deerflow.community.brave.tools import web_search_tool
result = web_search_tool.invoke({"query": "test"})
parsed = json.loads(result)
assert "error" in parsed
def test_sends_correct_headers_and_params(self, mock_config_with_key):
results = [{"title": "T", "url": "https://x.com", "description": "D"}]
mock_resp = _make_brave_response(results)
with patch("deerflow.community.brave.tools.httpx.Client") as mock_client_cls:
mock_get = mock_client_cls.return_value.__enter__.return_value.get
mock_get.return_value = mock_resp
from deerflow.community.brave.tools import web_search_tool
web_search_tool.invoke({"query": "hello world"})
call_kwargs = mock_get.call_args
headers = call_kwargs.kwargs["headers"]
params = call_kwargs.kwargs["params"]
assert headers["X-Subscription-Token"] == "test-brave-key"
assert params["q"] == "hello world"
assert params["count"] == 5
def test_uses_env_key_when_config_absent(self):
with patch("deerflow.community.brave.tools.get_app_config") as mock:
mock.return_value.get_tool_config.return_value = None
with patch.dict("os.environ", {"BRAVE_SEARCH_API_KEY": "env-only-key"}, clear=True):
results = [{"title": "T", "url": "https://x.com", "description": "D"}]
mock_resp = _make_brave_response(results)
with patch("deerflow.community.brave.tools.httpx.Client") as mock_client_cls:
mock_get = mock_client_cls.return_value.__enter__.return_value.get
mock_get.return_value = mock_resp
from deerflow.community.brave.tools import web_search_tool
web_search_tool.invoke({"query": "env key test"})
headers = mock_get.call_args.kwargs["headers"]
assert headers["X-Subscription-Token"] == "env-only-key"
def test_partial_fields_in_result(self, mock_config_with_key):
"""Missing title/url/description should default to empty string."""
results = [{}]
mock_resp = _make_brave_response(results)
with patch("deerflow.community.brave.tools.httpx.Client") as mock_client_cls:
mock_client_cls.return_value.__enter__.return_value.get.return_value = mock_resp
from deerflow.community.brave.tools import web_search_tool
result = web_search_tool.invoke({"query": "test"})
parsed = json.loads(result)
assert parsed["results"][0] == {"title": "", "url": "", "content": ""}
+32
View File
@@ -198,6 +198,38 @@ class TestCheckWebSearch:
assert result.fix is not None
assert "make setup" in result.fix
def test_brave_with_key_ok(self, tmp_path, monkeypatch):
monkeypatch.setenv("BRAVE_SEARCH_API_KEY", "bsa-test")
cfg = tmp_path / "config.yaml"
cfg.write_text("config_version: 5\ntools:\n - name: web_search\n use: deerflow.community.brave.tools:web_search_tool\n")
result = doctor.check_web_search(cfg)
assert result.status == "ok"
def test_brave_without_key_warns(self, tmp_path, monkeypatch):
monkeypatch.delenv("BRAVE_SEARCH_API_KEY", raising=False)
cfg = tmp_path / "config.yaml"
cfg.write_text("config_version: 5\ntools:\n - name: web_search\n use: deerflow.community.brave.tools:web_search_tool\n")
result = doctor.check_web_search(cfg)
assert result.status == "warn"
assert result.fix is not None
assert "BRAVE_SEARCH_API_KEY" in result.fix
def test_brave_with_inline_api_key_ok(self, tmp_path, monkeypatch):
monkeypatch.delenv("BRAVE_SEARCH_API_KEY", raising=False)
cfg = tmp_path / "config.yaml"
cfg.write_text('config_version: 5\ntools:\n - name: web_search\n use: deerflow.community.brave.tools:web_search_tool\n api_key: "inline-key"\n')
result = doctor.check_web_search(cfg)
assert result.status == "ok"
assert "api_key configured" in result.detail
def test_brave_with_api_key_env_ref_ok(self, tmp_path, monkeypatch):
monkeypatch.setenv("BRAVE_SEARCH_API_KEY", "bsa-test")
cfg = tmp_path / "config.yaml"
cfg.write_text("config_version: 5\ntools:\n - name: web_search\n use: deerflow.community.brave.tools:web_search_tool\n api_key: $BRAVE_SEARCH_API_KEY\n")
result = doctor.check_web_search(cfg)
assert result.status == "ok"
assert "api_key" in result.detail
def test_no_search_tool_warns(self, tmp_path):
cfg = tmp_path / "config.yaml"
cfg.write_text("config_version: 5\ntools: []\n")