Composio powers 1000+ toolkits, tool search, context management, authentication, and a sandboxed workbench to help you build AI agents that turn intent into action.
Mirrored from https://github.com/Camo67/composio.git
Synced Nov 22, 20:45 UTC
docs: add llamaindex provider page and update gpt-4-turbo to gpt-5 (#1986)
Co-authored-by: Musthaq Ahamad <53584487+haxzie@users.noreply.github.com> Co-authored-by: Soham <sohamganatra1@gmail.com>
Prathit committed
Sep 28, 2025 at 03:57 UTC
abc938df6181832bcc684277c80ca09be18803e3
13 files changed
+74
-11
fern/docs.yml
+3
-1
@@ -235,6 +235,8 @@ navigation:
235
path: pages/dist/providers/google.mdx
236
- page: LangChain Provider
237
path: pages/dist/providers/langchain.mdx
238
+ - page: LlamaIndex Provider
239
+ path: pages/dist/providers/llamaindex.mdx
240
- page: CrewAI Provider
241
path: pages/dist/providers/crewai.mdx
242
- page: Vercel AI SDK Provider
@@ -3377,4 +3379,4 @@ navigation:
3379
skip-slug: true
3380
paginated: false
3381
flattened: true
3380
-- tab: changelog
3382
+- tab: changelog
\ No newline at end of file
fern/pages/src/examples/utility_agents/gmail_labeler.mdx
+1
-1
@@ -47,7 +47,7 @@ def create_agent(user_id: str, composio_client: Composio[LangchainProvider]):
47
prompt = hub.pull("hwchase17/openai-functions-agent")
48
49
# Step 3: Initialize chat model.
50
- openai_client = ChatOpenAI(model="gpt-4-turbo")
50
+ openai_client = ChatOpenAI(model="gpt-5")
51
52
# Step 4: Define agent
53
return AgentExecutor(
fern/pages/src/examples/utility_agents/slack_summariser.mdx
+1
-1
@@ -45,7 +45,7 @@ def create_agent(user_id: str, composio_client: Composio[LangchainProvider]):
45
prompt = hub.pull("hwchase17/openai-functions-agent")
46
47
# Step 3: Initialize chat model.
48
- openai_client = ChatOpenAI(model="gpt-4-turbo")
48
+ openai_client = ChatOpenAI(model="gpt-5")
49
50
# Step 4: Define agent
51
return AgentExecutor(
fern/pages/src/examples/utility_agents/supabase_sql_agent.mdx
+1
-1
@@ -36,7 +36,7 @@ def create_agent(user_id: str, composio_client: Composio[LlamaIndexProvider]):
36
"""
37
38
# Setup client
39
- llm = OpenAI(model="gpt-4-turbo")
39
+ llm = OpenAI(model="gpt-5")
40
41
# Get All the tools
42
tools = composio_client.tools.get(
fern/pages/src/providers/llamaindex.mdx
new
+25
@@ -0,0 +1,25 @@
1
+---
2
+title: LlamaIndex Provider
3
+slug: /providers/llamaindex
4
+image: "https://og.composio.dev/api/og?title=LlamaIndex%20Provider" # image for socials
5
+keywords: ""
6
+hide-nav-links: false
7
+---
8
+
9
+The LlamaIndex Provider transforms Composio tools into a format compatible with LlamaIndex's function calling capabilities.
10
+
11
+## Setup
12
+<CodeGroup>
13
+```bash title="Python" for="python"
14
+pip install composio_llamaindex==0.8.0 llama-index
15
+```
16
+</CodeGroup>
17
+
18
+## Usage
19
+
20
+<CodeGroup>
21
+<SnippetCode
22
+ src="fern/snippets/providers/python/llamaindex-provider.py"
23
+ title="Python"
24
+/>
25
+</CodeGroup>
\ No newline at end of file
fern/snippets/providers/python/langchain-provider.py
+1
-1
@@ -8,7 +8,7 @@ from langchain_openai import ChatOpenAI
8
prompt = hub.pull("hwchase17/openai-functions-agent")
9
10
# Initialize tools.
11
-openai_client = ChatOpenAI(model="gpt-4-turbo")
11
+openai_client = ChatOpenAI(model="gpt-5")
12
13
composio = Composio(provider=LangchainProvider())
14
fern/snippets/providers/python/llamaindex-provider.py
new
+36
@@ -0,0 +1,36 @@
1
+import asyncio
2
+import dotenv
3
+from composio_llamaindex import LlamaIndexProvider
4
+from llama_index.core.agent.workflow import FunctionAgent
5
+from llama_index.llms.openai import OpenAI
6
+
7
+from composio import Composio
8
+
9
+# Load environment variables from .env
10
+dotenv.load_dotenv()
11
+
12
+# Setup client
13
+llm = OpenAI(model="gpt-5")
14
+composio = Composio(provider=LlamaIndexProvider())
15
+
16
+tools = composio.tools.get(
17
+ user_id="user@acme.com",
18
+ tools=["GITHUB_STAR_A_REPOSITORY_FOR_THE_AUTHENTICATED_USER"],
19
+)
20
+
21
+workflow = FunctionAgent(
22
+ tools=tools,
23
+ llm=llm,
24
+ system_prompt="You are an agent that performs github actions.",
25
+)
26
+
27
+
28
+async def main():
29
+ result = await workflow.run(
30
+ user_msg="Hello! I would like to star a repo composiohq/composio on GitHub"
31
+ )
32
+ print(result)
33
+
34
+
35
+if __name__ == "__main__":
36
+ asyncio.run(main())
\ No newline at end of file
python/providers/autogen/autogen_demo.py
+1
-1
@@ -22,7 +22,7 @@ def main():
22
system_message="Reply TERMINATE when the task is done or when user's content is empty",
23
llm_config={
24
"config_list": [
25
- {"model": "gpt-4-turbo", "api_key": os.environ["OPENAI_API_KEY"]},
25
+ {"model": "gpt-5", "api_key": os.environ["OPENAI_API_KEY"]},
26
]
27
},
28
)
python/providers/langchain/langchain_demo.py
+1
-1
@@ -13,7 +13,7 @@ from composio import Composio
13
prompt = hub.pull("hwchase17/openai-functions-agent")
14
15
# Initialize tools.
16
-openai_client = ChatOpenAI(model="gpt-4-turbo")
16
+openai_client = ChatOpenAI(model="gpt-5")
17
18
19
def main():
python/providers/llamaindex/llamaindex_demo.py
+1
-1
@@ -11,7 +11,7 @@ from composio import Composio
11
dotenv.load_dotenv()
12
13
# Setup client
14
-llm = OpenAI(model="gpt-4-turbo")
14
+llm = OpenAI(model="gpt-5")
15
composio = Composio(provider=LlamaIndexProvider())
16
17
# Get All the tools
python/providers/openai/README.md
+1
-1
@@ -52,7 +52,7 @@ This step involves configuring and executing the agent to carry out actions, suc
52
my_task = "Star a repo composiohq/composio on GitHub"
53
54
# Create a chat completion request to decide on the action
55
-response = openai_client.chat.completions.create(model="gpt-4-turbo-preview",
55
+response = openai_client.chat.completions.create(model="gpt-5",
56
tools=actions, # Passing actions we fetched earlier.
57
messages=[
58
{"role": "system", "content": "You are a helpful assistant."},
python/providers/openai/openai_assistant_demo.py
+1
-1
@@ -25,7 +25,7 @@ assistant_instruction = (
25
assistant = openai_client.beta.assistants.create(
26
name="Personal Assistant",
27
instructions=assistant_instruction,
28
- model="gpt-4-turbo-preview",
28
+ model="gpt-5",
29
tools=actions, # type: ignore
30
)
31
ts/docs/getting-started.md
+1
-1
@@ -160,7 +160,7 @@ const tools = await composio.tools.get('default', {
160
161
// Create a chat completion with OpenAI using the tools
162
const completion = await openai.chat.completions.create({
163
- model: 'gpt-4-turbo',
163
+ model: 'gpt-5',
164
messages: [
165
{ role: 'system', content: 'You are a helpful assistant with access to GitHub tools.' },
166
{ role: 'user', content: 'List the repositories in the Composio organization' },