Codefresh MCP. Manage CI/CD and Delivery Clusters from 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.
Codefresh allows your AI agent to manage your entire CI/CD lifecycle conversationally. List pipelines, check build status, trigger manual deploys, and monitor Kubernetes clusters—all without leaving your chat window.
You get full visibility into GitOps workflows and deployment targets directly from any compatible client.
What your AI agents can do
Get build execution details
Retrieves the detailed status and logs for a specific, existing build run.
Get my codefresh profile
Gets basic information about the user account connected to Codefresh.
Get pipeline configuration
Fetches detailed settings and configurations for a specific CI/CD pipeline.
See a list of every defined CI/CD pipeline within your account.
Get the current execution status and detailed logs for any specific build run.
Start a new, manual build for a specified pipeline, branch, or set of variables.
List all connected Kubernetes and delivery clusters to verify deployment targets.
Retrieve a list of shared environment contexts, including secrets and variables used by workflows.
Ask AI about this MCP
Supported MCP Clients
OAuth 2.0 Compatible Gemini Waiting for input…
Codefresh: 8 Tools for DevSecOps
These tools let you list pipelines, check build status, trigger deployments, and audit every cluster connected to Codefresh.
Make your AI actually useful.
Add this MCP to Claude, Cursor, or Windsurf and your AI stops guessing. It gets real tools to look things up, take action, and handle the stuff you keep doing by hand.
Start using Codefresh on Vinkius019d7576get build execution details
Retrieves the detailed status and logs for a specific, existing build run.
019d7576get my codefresh profile
Gets basic information about the user account connected to Codefresh.
019d7576get pipeline configuration
Fetches detailed settings and configurations for a specific CI/CD pipeline.
019d7576list codefresh builds
Retrieves a list of all recent builds that have run in the account.
019d7576list codefresh pipelines
Lists every active CI/CD pipeline configured in your Codefresh account.
019d7576list delivery clusters
Shows all connected Kubernetes and delivery clusters for monitoring purposes.
019d7576list shared contexts
Lists the shared environment contexts, including variables and secrets used in workflows.
019d7576trigger codefresh build
Starts a new build run for a specified pipeline using defined parameters.
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 Codefresh, then connect any of our 4,800+ other servers whenever your AI needs more. One click, no limits.
- Use this MCP plus 4,800+ 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
Independent Platform Disclaimer: Vinkius is an independent platform and is not affiliated with, endorsed by, sponsored by, verified by, or otherwise authorized by Codefresh. 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 8 capabilities that interface natively with Claude, ChatGPT, Cursor, and any MCP client. No middleware. No custom integration required.
Manually tracking deployments is a nightmare.
Right now, if you want to know what’s happening with your code, you open the Codefresh UI. You click on the pipeline name. Then you navigate to the build run. You check the status widget. If that fails, you have to find the error logs in a separate tab or section. It's clicking through five different pages just for one answer.
With this MCP, your agent handles it all. You talk about what needs fixing—say, 'What happened with the API service build?'—and the MCP gets the details from `get_build_execution_details` and pipes the status right into your chat. It’s immediate.
Check Build Status with get_build_execution_details
Before, checking a build meant logging in, finding the specific pipeline run, and then scrolling through potentially thousands of lines of text just to find the failure point. If you needed context on why it failed, you were dead in the water.
Now, your agent runs `get_build_execution_details` and gives you a summary: 'Failure at Step 3: Authentication token expired.' It cuts out all that noise and gives you exactly what you need to fix it.
What you can do with this MCP connector
Managing software deployments used to mean juggling dashboards: checking one page for pipeline health, another for cluster status, and a third just to find the right variable. This MCP changes that. Connect it to your agent, and you handle all of it via natural conversation.
You can ask about every active deployment pipeline or check the detailed status of a recent build without any clicks. Need to kick off a manual run? Just ask for it, specifying the branch or variables needed. If you need to know what secrets are used across workflows, this MCP lists those shared contexts too.
The system monitors all connected Kubernetes and delivery clusters so you always verify where code is going. Because we handle API keys through a zero-trust proxy, your credentials never sit on disk; they only move in transit, making the whole operation secure.
019d7576-69f4-71dc-822e-6c642638e28e 
How Codefresh MCP Works
- 1 Subscribe to the Codefresh MCP and input your API Key.
- 2 Connect this MCP to your agent client (Claude, Cursor, etc.).
- 3 Tell your agent what you need done with natural language commands.
The bottom line is: your AI agent handles all the calls, connecting Codefresh data directly into your chat interface.
Who Is Codefresh MCP For?
This is for the DevOps Engineer who gets tired of clicking through a dozen dashboards just to verify one deployment step. It's for the Release Manager who needs instant audit reports on cluster status, and any developer who wants build details without leaving their chat.
Uses this MCP to monitor pipeline health or trigger manual builds when a change hits staging.
Checks the status of all delivery clusters and audits deployment success rates without opening any web UI.
Quickly looks up build execution details or verifies environmental secrets straight from their terminal chat.
What Changes When You Connect
- Stop opening the dashboard to check status. Use this MCP to get build details instantly using
get_build_execution_detailsdirectly in your chat. - Never miss a deployment target. You can list all connected clusters via
list_delivery_clusters, confirming where code is actually going. - Need to debug? Instead of opening the config page, ask the agent to pull pipeline details using
get_pipeline_configuration. It’s faster. - Manual deploys happen in seconds. Use
trigger_codefresh_buildto kick off a build run with just natural language instructions. - Audit your environment easily. Running
list_shared_contextsgives you visibility into every secret and variable used across the system.
Real-World Use Cases
The deployment failed, but I don't know why.
I just need to check the last run. My agent uses list_codefresh_builds first, then runs get_build_execution_details on the failing item so I can immediately see the error logs and fix it.
We need to test a feature in production.
I don't want to wait for the scheduled run. I ask my agent to use trigger_codefresh_build, specifying the 'pre-release' branch, and it starts the process immediately.
The new microservice needs a deployment target.
Before writing any code, I confirm connectivity by asking for all clusters via list_delivery_clusters. This validates if our infrastructure is ready for the change.
I need to verify what variables are available for the new build.
Instead of digging through documentation, I ask my agent to run list_shared_contexts so I can confirm that 'database-endpoint' is actually defined for this environment.
The Tradeoffs
Over-reliance on the web dashboard
The dev team gets stuck in a loop: Dashboard -> Click Pipelines -> Check build status manually. This takes 5 minutes and requires multiple tabs.
→
Use your agent to run list_codefresh_pipelines first, then immediately ask for get_build_execution_details on the specific pipeline ID. It gets you the answer in two lines of chat.
Assuming build status is always visible
A developer asks about a build run that happened last month, but the system doesn't have an easy way to locate old metadata.
→
Use list_codefresh_builds first. This tool pulls all recent runs, giving your agent enough data to find the historical context you need.
Trying to guess required variables
A user tries to trigger a build without knowing if the necessary secrets are available in the current environment.
→
Always run list_shared_contexts before triggering. It shows exactly which variables and secrets your pipeline relies on, stopping guesswork.
When It Fits, When It Doesn't
Use this MCP if your primary job involves monitoring or initiating CI/CD workflows across multiple platforms (Kubernetes, GitOps). You need to check build histories (list_codefresh_builds) and manually trigger runs (trigger_codefresh_build) without opening a dedicated dashboard. Don't use it just because you want to see pipeline names; if you only need a simple list, most other basic automation tools cover that. This MCP shines when you combine its data—like checking the get_pipeline_configuration alongside list_delivery_clusters—to build a full picture of deployment readiness.
Common Questions About Codefresh MCP
How do I use `list_codefresh_pipelines` with this MCP? +
The agent calls list_codefresh_pipelines and sends back a list of every pipeline name. This lets you see what pipelines are available to manage or run builds on.
Does the Codefresh MCP let me check secrets using `list_shared_contexts`? +
Yes, running list_shared_contexts shows all shared environment contexts. This means you can see which variables and secrets are available for your build runs.
If I need to start a new deployment, do I use the MCP? +
Yes. To trigger a run, the agent uses trigger_codefresh_build. You just tell it which pipeline and what variables you want for the build.
What is the difference between listing builds and getting details using `list_codefresh_builds` vs. `get_build_execution_details`? +
list_codefresh_builds only gives you a list of recent runs (like titles). You need get_build_execution_details to pull the actual logs and status for one specific build.
How do I check my account details or verify connection using `get_my_codefresh_profile`? +
It retrieves current user and account information. This confirms the MCP has successfully authenticated against Codefresh without needing to access sensitive deployment data. You get basic metadata about your active credentials.
Before I trigger a build, how do I use `get_pipeline_configuration` to inspect its required settings? +
This function returns the complete schema and detailed definition of any specific pipeline. It lets you see variables or conditions that must be met before a build can run successfully.
If I need an audit of my deployment targets, how do I use `list_delivery_clusters`? +
It lists every connected Kubernetes and delivery cluster registered in your account. This is useful for confirming which environments are available to monitor or deploy against.
Are there limits if I run `list_codefresh_builds` repeatedly in a single session? +
The MCP handles large data sets efficiently, but standard API rate limits apply. If you encounter throttling errors, wait a minute or adjust your request to filter the list by date range.
Use it with your favorite AI tools
Connect this server to Cursor, Claude, VS Code, and more.