How to integrate Circleci MCP with LlamaIndex

Trusted by
AWS
Glean
Zoom
Airtable
GET STARTED FOR FREEGET A DEMO

30 min · no commitment · see it on your stack

Circleci logo
LlamaIndex logo
divider

Introduction

This guide walks you through connecting Circleci to LlamaIndex using the Composio tool router. By the end, you'll have a working Circleci agent that can trigger a new pipeline on main branch, list all pipelines for backend service, get test results from last successful build through natural language commands.

This guide will help you understand how to give your LlamaIndex agent real control over a Circleci account through Composio's Circleci MCP server.

Before we dive in, let's take a quick look at the key ideas and tools involved.

Also integrate Circleci with

TL;DR

Here's what you'll learn:
  • Set your OpenAI and Composio API keys
  • Install LlamaIndex and Composio packages
  • Create a Composio Tool Router session for Circleci
  • Connect LlamaIndex to the Circleci MCP server
  • Build a Circleci-powered agent using LlamaIndex
  • Interact with Circleci through natural language

What is LlamaIndex?

LlamaIndex is a data framework for building LLM applications. It provides tools for connecting LLMs to external data sources and services through agents and tools.

Key features include:

  • ReAct Agent: Reasoning and acting pattern for tool-using agents
  • MCP Tools: Native support for Model Context Protocol
  • Context Management: Maintain conversation context across interactions
  • Async Support: Built for async/await patterns

What is the Circleci MCP server, and what's possible with it?

The Circleci MCP server is an implementation of the Model Context Protocol that connects your AI agent and assistants like Claude, Cursor, etc directly to your Circleci account. It provides structured and secure access to your Circleci projects and pipelines, so your agent can trigger builds, fetch job artifacts, monitor workflows, and analyze test results on your behalf.

  • Automated pipeline triggering and management: Let your agent start new builds for specific branches or tags, enabling continuous integration workflows without manual intervention.
  • Workflow and job status monitoring: Ask your agent to fetch detailed information about jobs and workflows, including status, timing, and execution environment, to stay on top of your CI/CD processes.
  • Artifact and test result retrieval: Have the agent collect job artifacts or extract comprehensive test metadata and failure messages for easier debugging and reporting.
  • Pipeline and runner insights: Get your agent to list all pipelines for a project or enumerate available self-hosted runners, making it simple to manage and audit your Circleci resources.
  • User and configuration access: Retrieve user profile details or fetch pipeline YAML configurations as needed for documentation, troubleshooting, or workflow optimization.

Supported Tools & Triggers

Tools
Create ContextTool to create a new context in CircleCI.
Create Context (GraphQL)Tool to create a new CircleCI context using the GraphQL API.
Create Context RestrictionTool to create a context restriction in CircleCI.
Create Organization Orb AllowlistTool to create a new URL Orb allow-list entry for an organization.
Create Organization ProjectTool to create a new project within a CircleCI organization.
Create Organization GroupTool to create a group in an organization.
Create Project Environment VariableTool to create a new environment variable for a CircleCI project.
Create Usage Export JobTool to create a usage export job for a CircleCI organization.
Delete Context (GraphQL)Tool to delete a CircleCI context by its UUID using GraphQL API.
Delete Context RestrictionTool to delete a context restriction by its ID.
Delete Namespace and Related OrbsTool to delete a CircleCI registry namespace and all its associated orbs.
Delete Namespace AliasTool to remove a namespace alias by name in CircleCI.
Delete Organization Orb Allowlist EntryTool to remove an entry from the organization's URL orb allow-list.
Delete Organization GroupTool to delete a group from a CircleCI organization.
Delete ProjectTool to delete a CircleCI project and its settings.
Delete Project Environment VariableTool to delete an environment variable from a CircleCI project.
Get ContextTool to retrieve a context by its unique ID.
Get Current UserTool to retrieve information about the currently authenticated user.
Get Flaky TestsTool to get flaky tests for a project.
Get Job ArtifactsRetrieves artifacts (output files like test results, logs, build binaries, reports) produced by a CircleCI job.
Get Job DetailsTool to fetch details of a specific job within a project.
Get Orb DetailsTool to query detailed information about a CircleCI orb using the GraphQL API.
Get Orb VersionTool to retrieve detailed information about a specific CircleCI orb version via GraphQL.
Get OrganizationTool to retrieve organization details from CircleCI using GraphQL query.
Get Organization GroupTool to retrieve a group in an organization.
Get Pipeline ConfigTool to fetch pipeline configuration by ID.
Get Pipeline DefinitionTool to retrieve a pipeline definition by project and definition ID.
Get ProjectTool to retrieve a CircleCI project by its slug.
Get Project WorkflowsTool to get summary metrics for all workflows of a project.
Get Test MetadataTool to fetch test metadata for a specific job.
Get Usage Export JobTool to retrieve a usage export job by organization ID and job ID.
Get User InformationTool to retrieve information about a CircleCI user by their unique ID.
Get Workflow SummaryTool to get metrics and trends for a workflow.
List Context Environment VariablesTool to list all environment variables for a specific context.
List Insights BranchesTool to get all branches for a project from CircleCI Insights.
List Insights SummaryTool to get summary metrics with trends for the entire organization and for each project.
List Namespace OrbsTool to list orbs in a CircleCI registry namespace with pagination support.
List Orb CategoriesTool to retrieve all CircleCI orb categories with pagination support.
List OrbsTool to list CircleCI orbs with pagination support via GraphQL API.
List Organization GroupsTool to list all groups in a CircleCI organization.
List Pages SummaryTool to get summary metrics and trends for a project across its workflows and branches.
List Pipeline DefinitionsTool to list all pipeline definitions for a specific project.
List PipelinesTool to get a list of pipelines for an organization.
List Pipelines for ProjectTool to list all pipelines for a specific project.
List Project Environment VariablesTool to list all environment variables for a CircleCI project.
List Project SchedulesTool to list all schedules for a specific project.
List Self-Hosted RunnersList self-hosted runners in CircleCI.
List User CollaborationsTool to retrieve organizations where the authenticated user has access.
List Workflows by Pipeline IDTool to list all workflows associated with a specific pipeline.
List Workflows Jobs WorkflowsTool to get summary metrics for a project workflow's jobs.
List Workflows Test MetricsTool to get test metrics for a project's workflows.
Query ContextTool to retrieve a CircleCI context by its UUID using GraphQL API.
Query Namespace ExistsTool to determine if a namespace exists in the CircleCI registry.
Query Orb Category IDTool to fetch the unique category ID for a CircleCI orb category by its name.
Query Orb ExistsTool to check if an orb exists in CircleCI registry and retrieve its privacy status.
Query Orb IDTool to fetch an orb's ID and optionally its namespace ID by orb name.
Query Orb Latest VersionTool to fetch the latest published version of a CircleCI orb.
Query Orb SourceTool to retrieve source code of a specific CircleCI orb version via GraphQL.
Query Plan MetricsTool to query plan metrics including credit usage by project and organization for a date range.
Remove Context Environment Variable (GraphQL)Tool to remove an environment variable from a CircleCI context using GraphQL API.
Rename NamespaceTool to rename a CircleCI namespace by its UUID identifier.
Store Environment VariableTool to store an environment variable in a CircleCI context using GraphQL mutation.
Trigger PipelineTriggers a new CI/CD pipeline run for a specified CircleCI project.
Upsert Context Environment VariableTool to add or update an environment variable in a CircleCI context.
Validate Orb ConfigTool to validate CircleCI orb YAML configuration using the orbConfig GraphQL query.

What is the Composio tool router, and how does it fit here?

What is Composio SDK?

Composio's Composio SDK helps agents find the right tools for a task at runtime. You can plug in multiple toolkits (like Gmail, HubSpot, and GitHub), and the agent will identify the relevant app and action to complete multi-step workflows. This can reduce token usage and improve the reliability of tool calls. Read more here: Getting started with Composio SDK

The tool router generates a secure MCP URL that your agents can access to perform actions.

How the Composio SDK works

The Composio SDK follows a three-phase workflow:

  1. Discovery: Searches for tools matching your task and returns relevant toolkits with their details.
  2. Authentication: Checks for active connections. If missing, creates an auth config and returns a connection URL via Auth Link.
  3. Execution: Executes the action using the authenticated connection.

Step-by-step Guide

Prerequisites

Before you begin, make sure you have:
  • Python 3.8/Node 16 or higher installed
  • A Composio account with the API key
  • An OpenAI API key
  • A Circleci account and project
  • Basic familiarity with async Python/Typescript

Getting API Keys for OpenAI, Composio, and Circleci

OpenAI API key (OPENAI_API_KEY)
  • Go to the OpenAI dashboard
  • Create an API key if you don't have one
  • Assign it to OPENAI_API_KEY in .env
Composio API key and user ID
  • Log into the Composio dashboard
  • Copy your API key from Settings
    • Use this as COMPOSIO_API_KEY
  • Pick a stable user identifier (email or ID)
    • Use this as COMPOSIO_USER_ID

Installing dependencies

pip install composio-llamaindex llama-index llama-index-llms-openai llama-index-tools-mcp python-dotenv

Create a new Python project and install the necessary dependencies:

  • composio-llamaindex: Composio's LlamaIndex integration
  • llama-index: Core LlamaIndex framework
  • llama-index-llms-openai: OpenAI LLM integration
  • llama-index-tools-mcp: MCP client for LlamaIndex
  • python-dotenv: Environment variable management

Set environment variables

bash
OPENAI_API_KEY=your-openai-api-key
COMPOSIO_API_KEY=your-composio-api-key
COMPOSIO_USER_ID=your-user-id

Create a .env file in your project root:

These credentials will be used to:

  • Authenticate with OpenAI's GPT-5 model
  • Connect to Composio's Tool Router
  • Identify your Composio user session for Circleci access

Import modules

import asyncio
import os
import dotenv

from composio import Composio
from composio_llamaindex import LlamaIndexProvider
from llama_index.core.agent.workflow import ReActAgent
from llama_index.core.workflow import Context
from llama_index.llms.openai import OpenAI
from llama_index.tools.mcp import BasicMCPClient, McpToolSpec

dotenv.load_dotenv()

Create a new file called circleci_llamaindex_agent.py and import the required modules:

Key imports:

  • asyncio: For async/await support
  • Composio: Main client for Composio services
  • LlamaIndexProvider: Adapts Composio tools for LlamaIndex
  • ReActAgent: LlamaIndex's reasoning and action agent
  • BasicMCPClient: Connects to MCP endpoints
  • McpToolSpec: Converts MCP tools to LlamaIndex format

Load environment variables and initialize Composio

OPENAI_API_KEY = os.getenv("OPENAI_API_KEY")
COMPOSIO_API_KEY = os.getenv("COMPOSIO_API_KEY")
COMPOSIO_USER_ID = os.getenv("COMPOSIO_USER_ID")

if not OPENAI_API_KEY:
    raise ValueError("OPENAI_API_KEY is not set in the environment")
if not COMPOSIO_API_KEY:
    raise ValueError("COMPOSIO_API_KEY is not set in the environment")
if not COMPOSIO_USER_ID:
    raise ValueError("COMPOSIO_USER_ID is not set in the environment")

What's happening:

This ensures missing credentials cause early, clear errors before the agent attempts to initialise.

Create a Tool Router session and build the agent function

async def build_agent() -> ReActAgent:
    composio_client = Composio(
        api_key=COMPOSIO_API_KEY,
        provider=LlamaIndexProvider(),
    )

    session = composio_client.create(
        user_id=COMPOSIO_USER_ID,
        toolkits=["circleci"],
    )

    mcp_url = session.mcp.url
    print(f"Composio MCP URL: {mcp_url}")

    mcp_client = BasicMCPClient(mcp_url, headers={"x-api-key": COMPOSIO_API_KEY})
    mcp_tool_spec = McpToolSpec(client=mcp_client)
    tools = await mcp_tool_spec.to_tool_list_async()

    llm = OpenAI(model="gpt-5")

    description = "An agent that uses Composio Tool Router MCP tools to perform Circleci actions."
    system_prompt = """
    You are a helpful assistant connected to Composio Tool Router.
    Use the available tools to answer user queries and perform Circleci actions.
    """
    return ReActAgent(tools=tools, llm=llm, description=description, system_prompt=system_prompt, verbose=True)

What's happening here:

  • We create a Composio client using your API key and configure it with the LlamaIndex provider
  • We then create a tool router MCP session for your user, specifying the toolkits we want to use (in this case, circleci)
  • The session returns an MCP HTTP endpoint URL that acts as a gateway to all your configured tools
  • LlamaIndex will connect to this endpoint to dynamically discover and use the available Circleci tools.
  • The MCP tools are mapped to LlamaIndex-compatible tools and plug them into the Agent.

Create an interactive chat loop

async def chat_loop(agent: ReActAgent) -> None:
    ctx = Context(agent)
    print("Type 'quit', 'exit', or Ctrl+C to stop.")

    while True:
        try:
            user_input = input("\nYou: ").strip()
        except (KeyboardInterrupt, EOFError):
            print("\nBye!")
            break

        if not user_input or user_input.lower() in {"quit", "exit"}:
            print("Bye!")
            break

        try:
            print("Agent: ", end="", flush=True)
            handler = agent.run(user_input, ctx=ctx)

            async for event in handler.stream_events():
                # Stream token-by-token from LLM responses
                if hasattr(event, "delta") and event.delta:
                    print(event.delta, end="", flush=True)
                # Show tool calls as they happen
                elif hasattr(event, "tool_name"):
                    print(f"\n[Using tool: {event.tool_name}]", flush=True)

            # Get final response
            response = await handler
            print()  # Newline after streaming
        except KeyboardInterrupt:
            print("\n[Interrupted]")
            continue
        except Exception as e:
            print(f"\nError: {e}")

What's happening here:

  • We're creating a direct terminal interface to chat with your Circleci database
  • The LLM's responses are streamed to the CLI for faster interaction.
  • The agent uses context to maintain conversation history
  • You can type 'quit' or 'exit' to stop the chat loop gracefully
  • Agent responses and any errors are displayed in a clear, readable format

Define the main entry point

async def main() -> None:
    agent = await build_agent()
    await chat_loop(agent)

if __name__ == "__main__":
    # Handle Ctrl+C gracefully
    signal.signal(signal.SIGINT, lambda s, f: (print("\nBye!"), exit(0)))
    try:
        asyncio.run(main())
    except KeyboardInterrupt:
        print("\nBye!")

What's happening here:

  • We're orchestrating the entire application flow
  • The agent gets built with proper error handling
  • Then we kick off the interactive chat loop so you can start talking to Circleci

Run the agent

npx ts-node llamaindex-agent.ts

When prompted, authenticate and authorise your agent with Circleci, then start asking questions.

Complete Code

Here's the complete code to get you started with Circleci and LlamaIndex:

import asyncio
import os
import signal
import dotenv

from composio import Composio
from composio_llamaindex import LlamaIndexProvider
from llama_index.core.agent.workflow import ReActAgent
from llama_index.core.workflow import Context
from llama_index.llms.openai import OpenAI
from llama_index.tools.mcp import BasicMCPClient, McpToolSpec

dotenv.load_dotenv()

OPENAI_API_KEY = os.getenv("OPENAI_API_KEY")
COMPOSIO_API_KEY = os.getenv("COMPOSIO_API_KEY")
COMPOSIO_USER_ID = os.getenv("COMPOSIO_USER_ID")

if not OPENAI_API_KEY:
    raise ValueError("OPENAI_API_KEY is not set")
if not COMPOSIO_API_KEY:
    raise ValueError("COMPOSIO_API_KEY is not set")
if not COMPOSIO_USER_ID:
    raise ValueError("COMPOSIO_USER_ID is not set")

async def build_agent() -> ReActAgent:
    composio_client = Composio(
        api_key=COMPOSIO_API_KEY,
        provider=LlamaIndexProvider(),
    )

    session = composio_client.create(
        user_id=COMPOSIO_USER_ID,
        toolkits=["circleci"],
    )

    mcp_url = session.mcp.url
    print(f"Composio MCP URL: {mcp_url}")

    mcp_client = BasicMCPClient(mcp_url, headers={"x-api-key": COMPOSIO_API_KEY})
    mcp_tool_spec = McpToolSpec(client=mcp_client)
    tools = await mcp_tool_spec.to_tool_list_async()

    llm = OpenAI(model="gpt-5")
    description = "An agent that uses Composio Tool Router MCP tools to perform Circleci actions."
    system_prompt = """
    You are a helpful assistant connected to Composio Tool Router.
    Use the available tools to answer user queries and perform Circleci actions.
    """
    return ReActAgent(
        tools=tools,
        llm=llm,
        description=description,
        system_prompt=system_prompt,
        verbose=True,
    );

async def chat_loop(agent: ReActAgent) -> None:
    ctx = Context(agent)
    print("Type 'quit', 'exit', or Ctrl+C to stop.")

    while True:
        try:
            user_input = input("\nYou: ").strip()
        except (KeyboardInterrupt, EOFError):
            print("\nBye!")
            break

        if not user_input or user_input.lower() in {"quit", "exit"}:
            print("Bye!")
            break

        try:
            print("Agent: ", end="", flush=True)
            handler = agent.run(user_input, ctx=ctx)

            async for event in handler.stream_events():
                # Stream token-by-token from LLM responses
                if hasattr(event, "delta") and event.delta:
                    print(event.delta, end="", flush=True)
                # Show tool calls as they happen
                elif hasattr(event, "tool_name"):
                    print(f"\n[Using tool: {event.tool_name}]", flush=True)

            # Get final response
            response = await handler
            print()  # Newline after streaming
        except KeyboardInterrupt:
            print("\n[Interrupted]")
            continue
        except Exception as e:
            print(f"\nError: {e}")

async def main() -> None:
    agent = await build_agent()
    await chat_loop(agent)

if __name__ == "__main__":
    # Handle Ctrl+C gracefully
    signal.signal(signal.SIGINT, lambda s, f: (print("\nBye!"), exit(0)))
    try:
        asyncio.run(main())
    except KeyboardInterrupt:
        print("\nBye!")

Conclusion

You've successfully connected Circleci to LlamaIndex through Composio's Tool Router MCP layer. Key takeaways:
  • Tool Router dynamically exposes Circleci tools through an MCP endpoint
  • LlamaIndex's ReActAgent handles reasoning and orchestration; Composio handles integrations
  • The agent becomes more capable without increasing prompt size
  • Async Python provides clean, efficient execution of agent workflows
You can easily extend this to other toolkits like Gmail, Notion, Stripe, GitHub, and more by adding them to the toolkits parameter.

How to build Circleci MCP Agent with another framework

OpenAI Agents SDK logo

OpenAI Agents SDK

Use Circleci MCP with OpenAI Agents SDK

Claude Agent SDK logo

Claude Agent SDK

Use Circleci MCP with Claude Agent SDK

Claude Code logo

Claude Code

Use Circleci MCP with Claude Code

Claude Cowork logo

Claude Cowork

Use Circleci MCP with Claude Cowork

Codex logo

Codex

Use Circleci MCP with Codex

OpenClaw logo

OpenClaw

Use Circleci MCP with OpenClaw

Hermes logo

Hermes

Use Circleci MCP with Hermes

CLI logo

CLI

Use Circleci MCP with CLI

Google ADK logo

Google ADK

Use Circleci MCP with Google ADK

LangChain logo

LangChain

Use Circleci MCP with LangChain

Vercel AI SDK logo

Vercel AI SDK

Use Circleci MCP with Vercel AI SDK

Mastra AI logo

Mastra AI

Use Circleci MCP with Mastra AI

CrewAI logo

CrewAI

Use Circleci MCP with CrewAI

FAQ

What are the differences in Tool Router MCP and Circleci MCP?

With a standalone Circleci MCP server, the agents and LLMs can only access a fixed set of Circleci tools tied to that server. However, with the Composio Tool Router, agents can dynamically load tools from Circleci and many other apps based on the task at hand, all through a single MCP endpoint.

Can I use Tool Router MCP with LlamaIndex?

Yes, you can. LlamaIndex fully supports MCP integration. You get structured tool calling, message history handling, and model orchestration while Tool Router takes care of discovering and serving the right Circleci tools.

Can I manage the permissions and scopes for Circleci while using Tool Router?

Yes, absolutely. You can configure which Circleci scopes and actions are allowed when connecting your account to Composio. You can also bring your own OAuth credentials or API configuration so you keep full control over what the agent can do.

How safe is my data with Composio Tool Router?

All sensitive data such as tokens, keys, and configuration is fully encrypted at rest and in transit. Composio is SOC 2 Type 2 compliant and follows strict security practices so your Circleci data and credentials are handled as safely as possible.

Used by agents from

Context
Letta
glean
HubSpot
Agent.ai
Altera
DataStax
Entelligence
Rolai
Context
Letta
glean
HubSpot
Agent.ai
Altera
DataStax
Entelligence
Rolai
Context
Letta
glean
HubSpot
Agent.ai
Altera
DataStax
Entelligence
Rolai

Never worry about agent reliability

We handle tool reliability, observability, and security so you never have to second-guess an agent action.

Get started for freeGet a demo↗