Harness MCP. Manage pipelines, executions, and secrets from your chat.
Gemini Works with every AI agent you already use
…and any MCP-compatible client
Just plug in your AI agents and start using Vinkius.
Harness. This MCP Server manages your entire CI/CD and DevOps workflow. You can list projects, trigger pipeline runs, check real-time execution status, and review platform audit logs—all from your AI client.
It gives you full control over software delivery pipelines without leaving your IDE.
What your AI agents can do
Execute pipeline
Starts the process for a specific pipeline defined in your project.
Get audit logs
Retrieves a log of all changes and actions taken within the Harness platform.
Get execution status
Provides the current status and detailed step-by-step progress of a specific pipeline run.
You can list available pipelines, get their definitions, and manually trigger a pipeline run using execute_pipeline.
Use get_execution_status to get real-time step-by-step details and the current status of any running deployment.
Retrieve platform audit logs with get_audit_logs to track every change made to the Harness environment for compliance.
List all projects, environments, secrets, and connected infrastructure (Git, Docker, etc.) to understand the full scope of your DevOps setup.
List all microservices and project environments using list_services and list_environments to map your entire application architecture.
Ask AI about this MCP
Supported MCP Clients
Gemini Waiting for input…
Harness MCP Server: 11 Tools for DevOps
These 11 tools allow your AI client to perform every core action needed to manage, monitor, and audit complex CI/CD pipelines within Harness.
019d75aeexecute pipeline
Starts the process for a specific pipeline defined in your project.
019d75aeget audit logs
Retrieves a log of all changes and actions taken within the Harness platform.
019d75aeget execution status
Provides the current status and detailed step-by-step progress of a specific pipeline run.
019d75aeget pipeline
Gets the full YAML definition and details for a specified pipeline.
019d75aelist connectors
Lists all external infrastructure sources like Git, Docker, or Kubernetes connected to your account.
019d75aelist environments
Lists the deployment environments (e.g., Staging, Production) defined for a project.
019d75aelist executions
Shows a list of past and active runs for a particular pipeline.
019d75aelist pipelines
Lists all defined pipelines within a specific project scope.
019d75aelist projects
Returns a list of all projects managed within the connected Harness organization.
019d75aelist secrets
Displays all secrets (passwords, keys, tokens) configured and available for a project.
019d75aelist services
Lists all microservices and application components defined within a project.
Choose How to Get Started
Build a custom MCP for your own tools, or connect a ready-made integration from our catalog.
Build Your Own
Turn any API into an MCP. Import a spec, define Agent Skills, or deploy with MCPFusion.
- Import from OpenAPI, Swagger, or YAML specs
- Create Agent Skills with progressive disclosure
- Deploy to edge with MCPFusion framework
- Built in DLP, auth, and compliance on every call
- Real time usage dashboard and cost metering
- Publish to catalog or keep private
Make Your AI Do More
Start with Harness, then connect any of our 4,700+ other servers whenever your AI needs more. One click, no limits.
- Use this MCP plus 4,700+ others, all in one place
- Add new capabilities to your AI anytime you want
- Every connection is secured and compliant automatically
- Track usage and costs across all your servers
- Works with Claude, ChatGPT, Cursor, and more
- New servers added to the catalog every week
What you can do with this MCP connector
You're running your whole software delivery lifecycle through your AI client, connecting it right to your Harness platform. This server gives your agent the tools it needs to act like a full-time Release Engineer, letting you manage pipelines, monitor deployments, and audit changes without ever leaving your IDE. You've got full control over your CI/CD stuff.
Manage Software Pipelines
To start, your agent can list all the pipelines defined in a project using list_pipelines, grab the full YAML definition for a specific workflow with get_pipeline, or kick off a deployment run right away using execute_pipeline. You can also list all projects in your organization with list_projects and view all the microservices and application components defined in a project using list_services.
Track Live Deployments
Need to know what's happening right now? Your agent checks the current status and gets step-by-step progress for any running deployment using get_execution_status. You can also see a list of all past and active runs for a pipeline using list_executions.
Audit System Changes
For compliance, your agent pulls detailed platform audit logs using get_audit_logs, tracking every change and action taken on the Harness environment. You can also check all the secrets—passwords, keys, tokens—configured for a project by calling list_secrets.
Inventory Project Assets
Understanding your setup is key. Your agent can list all deployment environments, like Staging or Production, using list_environments, and it can also show you all the external infrastructure sources connected to your account, such as Git, Docker, or Kubernetes, by running list_connectors. To map out your entire DevOps setup, your agent can list all projects and view all defined environments and microservices.
Check the Scope
When you need a full picture, your agent can list every project in your connected Harness organization with list_projects, and it can show you every microservice and application component defined within those projects using list_services. You can also see what all the secrets are available for a project with list_secrets, and you can check the full list of all pipelines available in a project scope using list_pipelines.
How Harness MCP Works
- 1 First, you subscribe to the server and input your Harness API Key, Account ID, and Organization ID.
- 2 Your AI client then calls the specific tool (e.g.,
list_pipelines) to gather the necessary data. - 3 The client returns the data to you, allowing you to ask follow-up questions or trigger the next action (e.g., 'Now run the 'Production Deploy' pipeline').
The bottom line is, your AI client talks to the Harness API for you, handling all the API calls and data formatting.
Who Is Harness MCP For?
This is for the DevOps Engineer who gets tired of clicking through ten different dashboards just to check a single deployment status. It’s for the SRE who needs to audit failures across multiple services. If you spend more time navigating UIs than actually coding, you need this.
Triggers specific pipelines and retrieves status updates without leaving the IDE or chat window.
Monitors execution failures and pulls detailed audit logs to maintain platform stability and troubleshoot complex issues.
Automates oversight across multiple projects, checking which services are ready for deployment and verifying necessary secrets are in place.
What Changes When You Connect
- Instantly check deployment status. Instead of navigating to the dashboard, use
get_execution_statusto get real-time status and step details for any pipeline run. You know exactly where a deployment failed. - Automate deployments. Trigger a full workflow using
execute_pipeline. You don't need to log into the UI to kick off a production release; just ask your agent to do it. - Maintain compliance records. Use
get_audit_logsto retrieve platform audit logs. This creates an immutable record of who made changes and when, which is critical for security and compliance checks. - Map your entire stack. Use
list_projectsandlist_servicesto get an immediate overview of every project and microservice in your organization. No more guessing what assets exist. - Verify required credentials. Before running anything, use
list_secretsto confirm that all necessary credentials and keys are configured and available for the target project. - Scope your work quickly. Use
list_pipelinesandlist_environmentsto narrow down the scope immediately. You can list all pipelines for a project, then check which environments are defined.
Real-World Use Cases
Debugging a Failed Production Deploy
The build fails in Staging. Instead of clicking through the project dashboard, you ask your agent to check the status using get_execution_status. The agent reports that Step 3 failed, and you use get_audit_logs to see which user last modified the deployment definition, solving the problem in seconds.
Onboarding to a New Project
A new engineer needs to understand the entire system. They ask the agent to run list_projects and list_services. The agent gives them a complete map of the organization's assets, saving days of manual discovery and setup.
Pre-flight Check for Release
Before deploying, you need to verify the environment. You run list_environments and check list_secrets to confirm that the target environment (e.g., Prod) is set up and that the necessary API keys are present for the pipeline to run.
Reviewing Compliance Changes
The security team asks who changed the deployment process last month. You run get_audit_logs to get a full, chronological record of every change, pinpointing the exact time and user account responsible.
The Tradeoffs
Manual Dashboard Crawling
Logging into the Harness UI, clicking into the project, then selecting the pipeline, then finding the execution history, and finally checking the status step-by-step. It's slow, and you lose context.
→
Ask your agent directly: 'What's the status of the 'Production Deploy' pipeline for app_502?' The agent uses get_execution_status and gives you the answer immediately.
Assuming Credentials Exist
Attempting to trigger a deployment run without first knowing if the necessary secrets are configured, leading to a run failure halfway through.
→
First, use list_secrets to verify all required credentials. Then, use list_environments to confirm the target scope before calling execute_pipeline.
Confusing Projects with Services
Trying to find a microservice by searching the project list, only to realize the service lives in a separate, unlisted definition.
→
Always start by running list_services within the project context. This tool provides a clean list of all application components defined in the system.
Ignoring the Audit Trail
When a deployment goes wrong, assuming the code or the pipeline definition is the issue, without checking who last changed it.
→
Always run get_audit_logs first. It tells you who made the last change, narrowing the investigation immediately.
When It Fits, When It Doesn't
Use this if you need to manage the full lifecycle of a software delivery pipeline—from discovery and secret management to triggering and auditing. You need to know who did what and when across multiple projects and environments. Don't use this if you only need to list basic projects; use list_projects. Don't use this if you just need to check a single file's contents; use a specialized document retrieval tool. This server is for the operational state of your entire CI/CD system.
Independent Platform Disclaimer: Vinkius is an independent platform and is not affiliated with, endorsed by, sponsored by, verified by, or otherwise authorized by Harness. All third-party trademarks, logos, and brand names are the property of their respective owners. Their use on this website is strictly for informational purposes to identify service compatibility and interoperability.
VINKIUS INFRASTRUCTURE
Cloud Hosted
Managed infra
V8 Isolated
Sandboxed per request
Zero-Trust Proxy
No stored credentials
DLP Enforced
Policy on every call
GDPR Compliant
EU data residency
Token Compression
~60% cost reduction
Works with Claude, ChatGPT, Cursor, and more
The Model Context Protocol standardizes how applications expose capabilities to LLMs. Instead of operating in isolation, your AI gains direct access to external platforms, live data, and real-world actions through secure, standardized connections.
This server provides 11 capabilities that interface natively with Claude, ChatGPT, Cursor, and any MCP client. No middleware. No custom integration required.
Available Capabilities
Debugging a deployment failure shouldn't require switching between four different tabs.
Today, finding out why a deployment failed is a pain. You have to log into the CI/CD dashboard, find the project, locate the pipeline, find the specific run, and then drill down into the logs to see the step failure. It's a copy-paste nightmare just to get a status update.
With the Harness MCP Server, you ask your agent: 'Show me the status of the Production Deploy pipeline.' The agent runs `get_execution_status` and spits out the exact failure point, complete with step details. You get the answer immediately, right in your chat.
Harness MCP Server: Get full visibility into your infrastructure.
Before you can run a pipeline, you need to know what exists. You have to manually check if the project is listed (`list_projects`), if the environment is ready (`list_environments`), and if the API keys are set up (`list_secrets`). These are separate, manual checks that take time.
The agent handles the sequence. You ask it to check the deployment readiness, and it runs `list_projects`, then `list_environments`, and finally `list_secrets`—all in one flow. It gives you a complete, verified picture of the system's state.
Common Questions About Harness MCP
How do I use the `get_audit_logs` tool? +
The get_audit_logs tool retrieves a full log of every action taken on the Harness platform. You pass the required date range or user ID, and the server returns a structured record of platform changes.
What is the difference between `list_pipelines` and `list_projects`? +
You must first use list_projects to see all the containers (projects). Then, you use list_pipelines to see the specific workflows defined inside one of those containers.
Can I trigger a pipeline using `execute_pipeline`? +
Yes. The execute_pipeline tool takes the pipeline ID and project ID and initiates a deployment run. You just need to specify the correct target environment.
How do I check the status of a running deployment using `get_execution_status`? +
You call get_execution_status with the execution ID. It returns real-time progress, listing which steps are complete and which steps are currently running.
Do I need to run `list_connectors` first? +
No. list_connectors only shows you what external systems (like Git) are connected. You only run it if you need to verify the available sources of code or infrastructure.
How do I use `list_secrets` to check for necessary infrastructure credentials? +
It lists all configured secrets for a project. This is useful for checking if your build agent has the required credentials (like API keys or passwords) before starting a deployment.
What data does `get_pipeline` return, and how can I use it? +
It returns the full YAML definition and details for a specific pipeline. You can use this data to inspect the steps, validate the workflow, or modify the pipeline definition before triggering it.
How does `list_services` help me manage my microservices in a project? +
It gives you a list of all defined microservices within a project. Use this list to confirm the names and scope of the services your pipeline needs to deploy or manage.
How do I find my Harness API Key? +
Log in to Harness, go to My Profile (bottom left), and select API Keys. You can generate either a Personal API Key or use a Service Account Token from there.
Where can I find my Account ID and Org ID? +
The Account ID is visible in your browser's URL when logged in (e.g., account/XYZ123). The Organization ID is typically found in the project settings or sidebar (default is often 'default').
Can I trigger a pipeline execution via this integration? +
Yes! Use the execute_pipeline tool by providing the project ID and pipeline ID. Harness will initiate the workflow immediately.
Is the integration secure for managing secrets? +
Absolutely. The integration only lists the metadata of your secrets (like names and identifiers) through your API key. Your credentials are encrypted and stored securely in the Vinkius Cloud.
Use it with your favorite AI tools
Connect this server to Cursor, Claude, VS Code, and more.
More in this category
QuickNode
Manage blockchain infrastructure via QuickNode — create data streams, configure webhooks, and interact with RPC nodes directly from your AI agent.
BlazeMeter
Automate continuous performance testing via BlazeMeter — manage workspaces, trigger load tests, and analyze active run metrics securely via AI.
Equinix Metal
Manage Equinix Metal API keys, email settings, and project invitations directly from your AI agent.
You might also like
Procore
Manage construction projects via Procore — track RFIs, submittals, punch lists, daily logs, and drawings from any AI agent.
Learn Amp
Combine learning, engagement, and performance in one people development platform that helps employees grow and organizations thrive.
Crawlbase
Scrape and crawl via Crawlbase — perform HTML extraction, handle JS-rendered pages, bypass CAPTCHAs, and scrape social profiles directly from any AI agent.