admin-valentin/arcade-mcp
1
0
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity Actions

Return media attachments metadata when retrieving X tweets (#191)

Browse source 
Modifies X tweet tools to return metadata about media attachments
(photo, GIF or video) when retrieving a tweet by ID, username or
keywords.

The tool will always return media attachments by default. Since it's
only metadata, it shouldn't add significant network overhead to existing
implementations of the tool.

My guess is more often than not people will want this info included.
When not needed, it doesn't hurt to include by default. It'd be annoying
to have to ask the LLM to include it every time they need.
This commit is contained in:
Renato Byrro 2025-01-08 16:36:43 -03:00 committed by GitHub
parent b2bdfe2459
commit cd1fb648bd
No known key found for this signature in database
GPG key ID: B5690EEEBB952194
5 changed files with 77 additions and 22 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

23
toolkits/x/arcade_x/tools/tweets.py
View file

@ -6,6 +6,7 @@ from arcade.sdk.auth import X
from arcade.sdk.errors import RetryableToolError
from arcade_x.tools.utils import (
expand_attached_media,
expand_long_tweet,
expand_urls_in_tweets,
get_headers_with_token,
@ -81,13 +82,13 @@ async def search_recent_tweets_by_username(
max(max_results, 10), 100
), # X API does not allow 'max_results' less than 10 or greater than 100
"next_token": next_token,
"expansions": "author_id",
"user.fields": "id,name,username,entities",
"tweet.fields": "entities,note_tweet",
}
params = remove_none_values(params)
params = expand_attached_media(remove_none_values(params))
url = (
"https://api.x.com/2/tweets/search/recent?"
"expansions=author_id&user.fields=id,name,username,entities&tweet.fields=entities,note_tweet"
)
url = f"{TWEETS_URL}/search/recent"
async with httpx.AsyncClient() as client:
response = await client.get(url, headers=headers, params=params, timeout=10)
@ -151,13 +152,13 @@ async def search_recent_tweets_by_keywords(
max(max_results, 10), 100
), # X API does not allow 'max_results' less than 10 or greater than 100
"next_token": next_token,
"expansions": "author_id",
"user.fields": "id,name,username,entities",
"tweet.fields": "entities,note_tweet",
}
params = remove_none_values(params)
params = expand_attached_media(remove_none_values(params))
url = (
"https://api.x.com/2/tweets/search/recent?"
"expansions=author_id&user.fields=id,name,username,entities&tweet.fields=entities,note_tweet"
)
url = f"{TWEETS_URL}/search/recent"
async with httpx.AsyncClient() as client:
response = await client.get(url, headers=headers, params=params, timeout=10)
@ -192,6 +193,8 @@ async def lookup_tweet_by_id(
"user.fields": "id,name,username,entities",
"tweet.fields": "entities,note_tweet",
}
params = expand_attached_media(params)
url = f"{TWEETS_URL}/{tweet_id}"
async with httpx.AsyncClient() as client:

21
toolkits/x/arcade_x/tools/utils.py
View file

@ -137,3 +137,24 @@ def remove_none_values(params: dict) -> dict:
A new dictionary with None values removed
"""
return {k: v for k, v in params.items() if v is not None}
def expand_attached_media(params: dict) -> dict:
"""
Include attached media metadata in the request parameters.
"""
params["expansions"] += ",attachments.media_keys"
params["tweet.fields"] += ",attachments"
params["media.fields"] = ",".join([
# media_key, url and type are returned by default, added here for clarity
"media_key",
"url",
"type",
"duration_ms",
"height",
"width",
"preview_image_url",
"alt_text",
"public_metrics",
])
return params

12
toolkits/x/evals/eval_x_tools.py
View file

@ -80,9 +80,7 @@ def x_eval_suite() -> EvalSuite:
expected_tool_calls=[
ExpectedToolCall(
func=post_tweet,
args={
"tweet_text": "Hello World! Exciting stuff is happening over at Arcade AI!",
},
args={"tweet_text": "Hello World! Exciting stuff is happening over at Arcade AI!"},
)
],
critics=[
@ -139,7 +137,7 @@ def x_eval_suite() -> EvalSuite:
"max_results": 42,
"next_token": "b26v89c19zqg8o3frr3tekall7a7ooom3sctaw30rz62l",
},
)
),
],
critics=[
BinaryCritic(
@ -164,7 +162,7 @@ def x_eval_suite() -> EvalSuite:
ExpectedToolCall(
func=lookup_single_user_by_username,
args={"username": "jack"},
)
),
],
critics=[
BinaryCritic(
@ -186,7 +184,7 @@ def x_eval_suite() -> EvalSuite:
"phrases": ["Arcade AI"],
"max_results": 10,
},
)
),
],
critics=[
BinaryCritic(
@ -208,7 +206,7 @@ def x_eval_suite() -> EvalSuite:
ExpectedToolCall(
func=lookup_tweet_by_id,
args={"tweet_id": "123456789"},
)
),
],
critics=[
BinaryCritic(

2
toolkits/x/pyproject.toml
View file

@ -1,6 +1,6 @@
[tool.poetry]
name = "arcade_x"
version = "0.1.8"
version = "0.1.9"
description = "LLM tools for interacting with X (Twitter)"
authors = ["Arcade AI <dev@arcade-ai.com>"]

41
toolkits/x/tests/test_tweets.py
View file

@ -117,7 +117,12 @@ async def test_search_recent_tweets_by_username_success(tool_context, mock_httpx
},
}
],
"includes": {"users": [{"id": "0987654321", "name": "Test User", "username": "testuser"}]},
"includes": {
"users": [{"id": "0987654321", "name": "Test User", "username": "testuser"}],
"media": [
{"media_key": "1234567890", "type": "photo", "url": "https://example.com/photo.jpg"}
],
},
}
mock_httpx_client.get.return_value = mock_response
@ -127,6 +132,12 @@ async def test_search_recent_tweets_by_username_success(tool_context, mock_httpx
assert "data" in result
assert len(result["data"]) == 1
assert result["data"][0]["text"] == full_tweet_text
assert "includes" in result
assert "media" in result["includes"]
assert len(result["includes"]["media"]) == 1
assert result["includes"]["media"][0]["url"] == "https://example.com/photo.jpg"
mock_httpx_client.get.assert_called_once()
@ -168,16 +179,27 @@ async def test_search_recent_tweets_by_keywords_success(tool_context, mock_httpx
"entities": {},
}
],
"includes": {"users": [{"id": "0987654321", "name": "Test User", "username": "testuser"}]},
"includes": {
"users": [{"id": "0987654321", "name": "Test User", "username": "testuser"}],
"media": [
{"media_key": "1234567890", "type": "photo", "url": "https://example.com/photo.jpg"}
],
},
}
mock_httpx_client.get.return_value = mock_response
keywords = ["test", "keyword"]
result = await search_recent_tweets_by_keywords(tool_context, keywords=keywords)
result = await search_recent_tweets_by_keywords(context=tool_context, keywords=keywords)
assert "data" in result
assert len(result["data"]) == 1
assert result["data"][0]["text"] == full_tweet_text
assert "includes" in result
assert "media" in result["includes"]
assert len(result["includes"]["media"]) == 1
assert result["includes"]["media"][0]["url"] == "https://example.com/photo.jpg"
mock_httpx_client.get.assert_called_once()
@ -207,7 +229,12 @@ async def test_lookup_tweet_by_id_success(tool_context, mock_httpx_client):
},
"text": truncated_tweet_text,
"entities": {},
}
},
"includes": {
"media": [
{"media_key": "1234567890", "type": "photo", "url": "https://example.com/photo.jpg"}
]
},
}
mock_httpx_client.get.return_value = mock_response
@ -216,6 +243,12 @@ async def test_lookup_tweet_by_id_success(tool_context, mock_httpx_client):
assert "data" in result
assert result["data"]["text"] == full_tweet_text
assert "includes" in result
assert "media" in result["includes"]
assert len(result["includes"]["media"]) == 1
assert result["includes"]["media"][0]["url"] == "https://example.com/photo.jpg"
mock_httpx_client.get.assert_called_once()