Skip to main content

The #1 agentic semantic tool search: 91.6% first-try accuracy on S1 Search Bench Explore Tool Discovery

Live 82 Actions

CircleCI MCP Server
for AI Agents

Connect your AI agent to StackOne's CircleCI MCP server and give it 82 MCP tools out of the box. Auth, tool execution, and security all managed.

CircleCI logo
CircleCI MCP Server
Built by StackOne StackOne
DrataGPLocalyzeFlipMindtoolsScreenloop

Coverage

82 Agent Actions

Create, read, update, and delete across CircleCI — and extend your agent's capabilities with custom actions.

Authentication

Agent Tool Authentication

Per-user OAuth in one call. Your CircleCI MCP server gets session-scoped tokens with zero credentials stored on your infra.

Agent Auth →

Security

Agent Protection

Every CircleCI tool response scanned for prompt injection in milliseconds — 88.7% accuracy, all running on CPU.

Prompt Injection Defense →

Performance

Max Agent Context. Min Cost.

Free up to 96% of your agent's context window to enhance reasoning and reduce cost, on every CircleCI call.

Tools Discovery →

What is the CircleCI MCP Server?

A CircleCI MCP server lets AI agents read and write CircleCI data through the Model Context Protocol — Anthropic's open standard for connecting LLMs to external tools. StackOne's CircleCI MCP server ships with 82 pre-built actions, fully extensible via the Connector Builder — plus managed authentication, prompt injection defense, observability, and agent execution runtime. Connect it from MCP clients like Claude Desktop, Claude Code, Cursor, Goose, and VS Code, or from agent frameworks like OpenAI Agents SDK, LangChain, and Vercel AI SDK.

All CircleCI MCP Tools

Every action from CircleCI's API, ready for your agent. Create, read, update, and delete — scoped to exactly what you need.

Pipelines

  • List Pipelines

    Get a list of pipelines across ALL recently built projects in an organization (org-wide view, up to 250 projects). If you need pipelines for a SPECIFIC project, use list_project_pipelines instead. Requires org-slug from list_collaborations — never guess the slug format.

  • Get Pipeline

    Get a pipeline by ID

Contexts

  • Create Context

    Create a new context

  • List Contexts

    List all contexts for an owner (organization). Use owner-id (UUID) OR owner-slug (format "gh/org-name" for GitHub, "circleci/org-id" for GitLab/GitHub App). The owner-slug requires a VCS prefix — passing just the org name (e.g., "FalconTest") returns 404.

  • Get Context

    Get a context by ID

  • Delete Context

    Delete a context

Context Environment Variables

  • List Context Environment Variables

    List environment variables in a context

  • Delete Context Environment Variable

    Remove an environment variable from a context

Context Restrictions

  • Create Context Restriction

    Create a project or expression restriction on a context to control which projects can use it. Returns 409 Conflict if the restriction already exists — this means the restriction is already in place (treat as success). Returns 400 if the project ID or expression is invalid.

  • Get Context Restrictions

    Get a list of restrictions associated with a context

  • Delete Context Restriction

    Delete a restriction from a context

Organizations

  • Create Organization

    Create a new organization

  • Get Organization

    Get an organization by slug or ID

  • Delete Organization

    Delete an organization

Orb Allow-List Entrys

  • Create Orb Allow-List Entry

    Create a new URL Orb allow-list entry

  • Delete Orb Allow-List Entry

    Remove an entry from the org's URL orb allow-list

Webhooks

  • Create Webhook

    Create an outbound webhook

  • List Webhooks

    List outbound webhooks for a scope

  • Get Webhook

    Get a webhook by ID

  • Update Webhook

    Update an outbound webhook

  • Delete Webhook

    Delete an outbound webhook

Projects

  • Create Project

    Create a new project

  • Get Project

    Get a project by its project-slug (format "vcs-slug/org-name/repo-name"). Returns 404 if the project does not exist. IMPORTANT — never guess the project slug. First call list_collaborations to get the org slug, then call list_pipelines to find the exact project_slug from recent pipeline data.

Project Settings

  • Get Project Settings

    Get advanced settings for a project. The provider/organization/project path params are the three segments of the project slug — get the full slug from list_pipelines first, then split it.

  • Update Project Settings

    Update advanced settings for a project. The provider/organization/project path params are the three segments of the project slug — get the full slug from list_pipelines first, then split it.

Checkout Keys

  • Create Checkout Key

    Create a new checkout key for a project

  • List Checkout Keys

    Get all checkout keys for a project

Project Environment Variables

  • Create Project Environment Variable

    Create an environment variable for a project

  • List Project Environment Variables

    List all environment variables for a project

  • Get Project Environment Variable

    Get a masked environment variable by name

  • Delete Project Environment Variable

    Delete an environment variable from a project

Schedules

  • Create Schedule

    Create a pipeline schedule for a project

  • List Schedules

    List schedule triggers for a project

  • Get Schedule

    Get a schedule by ID

  • Update Schedule

    Update a pipeline schedule

  • Delete Schedule

    Delete a schedule

Pipeline Definitions

  • Create Pipeline Definition

    Create a pipeline definition for a project. The config_source.provider must match the project's actual integration type (e.g., use github_app if the project uses GitHub App, not github_server). Using a provider that doesn't match the project's configured integration returns 400. Check existing definitions via list_pipeline_definitions to see the correct provider.

  • List Pipeline Definitions

    List all pipeline definitions for a project

  • Get Pipeline Definition

    Get details for a pipeline definition

  • Update Pipeline Definition

    Update an existing pipeline definition's name, config file path, or checkout source. You cannot change the provider type to one the project doesn't support (e.g., switching from github_app to github_server returns 400 if no GitHub Server integration is configured). Use list_pipeline_definitions to see the current provider before updating.

  • Delete Pipeline Definition

    Delete a pipeline definition

Environments

  • List Environments

    List environments for an organization

  • Get Environment

    Get an environment by ID

Components

  • List Components

    List components for an organization

  • Get Component

    Get a component by ID

Groups

  • Create Group

    Create a new group in an organization

  • List Groups

    Get all groups in an organization

  • Get Group

    Get details for a group in an organization

  • Delete Group

    Delete a group in an organization

Other (33)

  • Add Or Update Context Environment Variable

    Add or update an environment variable in a context

  • Get Current User

    Get current authenticated user info

  • Get User

    Get a user by ID

  • List Collaborations

    List organizations the current user is a member of, including their org IDs and slugs. ALWAYS call this first before any project or org operation to get the correct org-slug and project-slug values. The slug format varies by provider — it may be "gh/org-name" for GitHub OAuth, "bb/org-name" for Bitbucket, or "circleci/org-id" for GitHub App and GitLab orgs. Never guess the slug format; always use the exact slug returned here.

  • List Project Pipelines

    Get all pipelines for a SPECIFIC project by its project-slug. Use this instead of list_pipelines when you want pipelines for one project. Get the project-slug from the project_slug field in the list_pipelines response — never guess it.

  • Get Pipeline Configuration

    Get the configuration for a pipeline

  • Get Pipeline Workflows

    Get all workflows for a pipeline

  • Get Pipeline By Number

    Get a pipeline by its pipeline number for a given project. Get the project-slug from the project_slug field in list_pipelines — never guess it.

  • Get Pipeline Values

    Get pipeline values for a pipeline

  • List My Pipelines

    Get all pipelines triggered by the current user for a project

  • List Orb Allow-List Entries

    List entries in an organization's URL Orb allow-list

  • Get Project Workflow Runs

    Get recent runs of a workflow

  • Get All Branches

    Get all branches for a project

  • Get Organization Summary Metrics

    Get aggregated summary metrics for an organization, including success rates, duration, and throughput across all projects. The optional project-names filter accepts project slug suffixes (the repo-name part of the project slug), NOT display names. Passing display names or invalid values causes a 400 error.

  • Get Project Workflow Metrics

    Get summary metrics for a project's workflows

  • Get Flaky Tests

    Get flaky tests for a project

  • Get Project Workflow Summary

    Get summary metrics and trends for a project across its workflows and branches

  • Get Job Timeseries Data

    Get timeseries data for all jobs within a workflow

  • Get Workflow

    Get a workflow by ID

  • Get Workflow Jobs

    Get all jobs for a workflow

  • Get Job Details

    Get job details by job number

  • Get Job Artifacts

    Get artifacts for a job

  • Get Job Test Metadata

    Get test metadata for a job

  • Get Job

    Get a job by ID

  • List Component Versions

    List versions for a component

  • Trigger Pipeline

    Trigger a new pipeline for a project

  • Trigger Pipeline Run

    Trigger a new pipeline run from a pipeline definition

  • Continue Pipeline

    Continue a pipeline from the setup phase

  • Approve Job

    Approve a pending approval job in a workflow

  • Cancel Workflow

    Cancel a running workflow

  • Rerun Workflow

    Rerun a workflow

  • Cancel Job

    Cancel a job by job number

  • Cancel Job By ID

    Cancel a job by its unique ID

Set Up Your CircleCI MCP Server in Minutes

One endpoint. Any framework. Your agent is talking to CircleCI in under 10 lines of code.

Agent Frameworks

Claude Desktop
{
  "mcpServers": {
    "stackone": {
      "command": "npx",
      "args": [
        "-y",
        "mcp-remote@latest",
        "https://api.stackone.com/mcp?x-account-id=<account_id>",
        "--header",
        "Authorization: Basic <YOUR_BASE64_TOKEN>"
      ]
    }
  }
}

Check More Developer Tools MCP Servers

Azure DevOps

185+ actions

Cloudflare

141+ actions

Linear

136+ actions

Bitbucket

133+ actions

Supabase

129+ actions

GitLab

125+ actions

Terraform

118+ actions

CircleCI MCP Server FAQ

Does StackOne have a CircleCI MCP server?
Yes. StackOne offers a hosted CircleCI MCP server with 82 pre-built actions, and every action is tested and QA'd by StackOne. Connect it to Claude, Cursor, and any other MCP client, or to any agent framework through the AI Action SDK. It ships with managed agent authentication, prompt injection defense, and tool discovery with server-side execution that preserve your agent's context window and keep reasoning performance.
CircleCI MCP server vs direct API integration — what's the difference?
A CircleCI MCP server and direct API integration serve different use cases. Direct API integration is for software-to-software — backend code calling CircleCI. A CircleCI MCP server is for AI agents — MCP clients like Claude and Cursor, plus framework agents built with OpenAI, LangChain, or Vercel AI — discovering and calling CircleCI at runtime. StackOne provides both.
How does CircleCI authentication work for AI agents?
CircleCI authentication for AI agents works through a StackOne Connect Session. Create one via the dashboard or the SDK — you get an auth link and ready-to-paste config for Claude Desktop, Cursor, and other MCP clients. Your user authenticates their own CircleCI account; StackOne handles token exchange, storage, and refresh. Credentials never reach the LLM, and each user is isolated via origin_owner_id.
Are CircleCI MCP tools vulnerable to prompt injection?
Yes — CircleCI MCP tools can be vulnerable to indirect prompt injection. Any tool that reads user-written content — documents, messages, tickets, records, or free-text fields — is a potential vector. StackOne Defender scans every tool response before it enters the agent's context — regex patterns in ~1ms, then a MiniLM classifier in ~4ms. 88.7% accuracy, CPU-only.
What is the context bloat of a CircleCI agent and how do I avoid it?
Context bloat happens when CircleCI tool schemas and API responses eat your CircleCI agent's memory, preventing it from reasoning effectively. A single CircleCI query can return a massive JSON response, and connecting multiple tools compounds the problem. Tools Discovery and Code Mode reduce context bloat — loading only relevant tools per query and keeping raw responses out of the agent's context.
Can I limit which actions my CircleCI agent can access?
Yes — you can limit which actions your CircleCI agent can access directly from the StackOne dashboard. Toggle actions on or off, or restrict them to specific accounts, with no code changes to your agent. Session tokens can be scoped to exact actions so if one leaks, exposure stays contained.
Can I create custom agent actions for my CircleCI MCP server?
Yes — you can create custom agent actions for your CircleCI MCP server using Connector Builder. It's an integration agent your coding assistant (Claude Code, Cursor, or Copilot) can invoke to research CircleCI's API, generate production-ready connector YAML, test against the live API, and validate before you ship.
When should I NOT use a CircleCI MCP server?
Skip a CircleCI MCP server if your integration is purely software-to-software — direct CircleCI API integration is simpler when no AI agent is involved. For deterministic, compliance-critical operations (financial transactions, regulatory reporting), direct API gives you predictable behavior without agent-driven decision-making. MCP shines when AI agents need to dynamically discover and call CircleCI actions at runtime.
What AI frameworks and AI clients does the StackOne CircleCI MCP server support?
The StackOne CircleCI MCP server supports both. MCP clients (paste-and-go apps): Claude Desktop, Claude Code, Cursor, VS Code, Goose. Agent frameworks (code SDKs you build with): OpenAI Agents SDK, Anthropic, Vercel AI, Google ADK, CrewAI, Pydantic AI, LangChain, LangGraph, Azure AI Foundry.

Put your AI agents to work

All the tools you need to build and scale AI agent integrations, with best-in-class connectivity, execution, and security.