GitHub MCP. Manage code, issues, and history via conversation.
Gemini Works with every AI agent you already use
…and any MCP-compatible client
Just plug in your AI agents and start using Vinkius.
GitHub MCP connects your AI agent directly to the world's largest developer platform. You manage repositories, track issues, and search code without ever leaving your chat interface.
Instantly list repos across organizations, check file contents in any project folder, or open new issues—all through natural conversation.
What your AI agents can do
Add issue comment
Adds a comment to an existing issue or pull request using standard markdown formatting.
Create new issue
Opens a brand new, structured bug report or feature request within a repository.
Create pull request
Generates and submits a pull request between two specific branches in a repository.
You can list all repositories for a user or organization and get metadata on specific projects to understand the overall scope.
The agent lets you open new issues, add comments to existing tickets, and create pull requests directly from your conversation.
You can retrieve the full contents of specific files or directories across any repository, allowing deep code context for analysis.
The MCP lets you list recent commits and track all open issues and pull requests to monitor project activity.
You can perform powerful searches across GitHub's database to isolate specific code snippets or find relevant repositories by name.
Ask AI about this MCP
Supported MCP Clients
OAuth 2.0 Compatible Gemini Waiting for input…
GitHub: 18 Tools for Development Workflows
Use these tools to manage everything from opening new issues and tracking commits to retrieving the contents of any file in a repository.
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 GitHub on Vinkius019e9a8eadd issue comment
Adds a comment to an existing issue or pull request using standard markdown formatting.
019d75a5create new issue
Opens a brand new, structured bug report or feature request within a repository.
019e9a8ecreate pull request
Generates and submits a pull request between two specific branches in a repository.
019d75a5get file contents
Reads and returns the full text content of any specified file or directory within a project.
019e9a8eget issue details
Retrieves all specific information about one issue, given its unique number.
019d75a5get my github profile
Returns detailed identity and permission context for the connected user account.
019d75a5get repository details
Retrieves essential metadata, like name and visibility, for a single specified repository.
019e9a8elist branches
Lists all active branches associated with a given repository.
019e9a8elist commits
Shows recent changes to the code, optionally filtering by branch name or SHA hash.
019d75a5list org repositories
Retrieves a list of all repositories belonging to an organization account.
019d75a5list pull requests
Gathers a summary list of current and past pull requests for a project.
019e9a8elist releases
Fetches information about all official releases tagged within the repository's history.
019d75a5list repo issues
Retrieves a list of open and closed issues belonging to a specific repository.
019e9a8elist stargazers
Lists the users who have starred the repository, useful for community engagement metrics.
019d75a5list user repositories
Retrieves a list of all repositories owned by the connected user account.
019d75a5search github code
Executes a targeted search across code snippets within your codebase for specific keywords or patterns.
019d75a5search github repositories
Searches all available repositories by name or keyword, regardless of ownership.
019d75a5verify api connection
Checks the connection status to ensure your authentication token is valid and active.
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 GitHub, then connect any of our 4,900+ other servers whenever your AI needs more. One click, no limits.
- Use this MCP plus 4,900+ 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 GitHub. 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 18 capabilities that interface natively with Claude, ChatGPT, Cursor, and any MCP client. No middleware. No custom integration required.
Juggling dashboards is exhausting.
Right now, if you need to know the status of a feature, you're clicking through tabs: checking the issues list for open bugs, then navigating to the PR section to see reviews, and finally pulling up the repository settings to check the latest commit. It’s three or four separate clicks just to get one snapshot.
With this MCP, you simply ask your agent, 'What's the status of Feature X?' Your AI client instantly executes multiple internal checks—listing issues, checking PR status via `list_pull_requests`, and retrieving metadata—and hands you a single, coherent answer. You just get results.
The GitHub MCP gives you actionable code context.
Manually comparing multiple files or tracking down the exact implementation of a feature requires opening several tabs and copy-pasting snippets back and forth. This is slow, error-prone work that breaks focus.
Now, if your agent needs to understand code, it uses `get_file_contents`. It pulls the entire text content directly into the conversation flow. That means complex analysis or cross-referencing files happens in minutes, not hours.
What you can do with this MCP connector
Need to keep up with complex codebases? This MCP gives your AI agent direct access to GitHub's entire ecosystem. Forget switching tabs and manually digging into dashboards just to figure out who changed what or why a feature request stalled. You can ask your agent to list all repositories for a specific organization, grab the full contents of a README file, or even track down every commit made in the last week.
It’s about getting answers fast. Instead of navigating through dozens of menus, you just talk to your AI client. Want to see if a pull request is ready to merge? Ask. Need to open a new bug report and tag three people? Tell it to do it. This capability makes GitHub's massive data set actionable right where you work.
It’s available via the Vinkius Marketplace, giving any MCP-compatible client deep, structured access to your development history.
019d75a5-5797-71cd-bcb0-5c00a4eb4d74 
How GitHub MCP Works
- 1 First, subscribe to this MCP via the Vinkius Marketplace and provide your GitHub Personal Access Token.
- 2 Next, connect that token to any compatible AI client. This authorizes your agent to read and write data on your behalf.
- 3 Finally, start by telling your agent what you need—for example, 'List all open issues for Project X'—and the tool executes the request.
The bottom line is: you authorize your AI client once, then it handles complex GitHub workflows via simple conversation prompts.
Who Is GitHub MCP For?
Anyone who spends time on a code dashboard and gets frustrated by clicking through tabs or manually copying IDs. It's for the developer tired of context switching and the project manager who needs an instant, high-level view of team health.
Needs to check a pull request status, read file contents, or list commits quickly so they can debug without leaving their primary workspace.
Wakes up needing an overview of repository activity and community engagement; they use the MCP to get real-time summaries of open issues.
Uses the tool to automate retrieving repository metadata or file contents for auditing purposes, keeping workflows consistent and documented.
What Changes When You Connect
- Instantly see the status of every project. You can use
list_org_repositoriesto list all projects owned by an organization in one go, instead of visiting multiple dashboard tabs. - Keep track of development progress with minimal effort. Use
list_pull_requestsandget_issue_detailsso your agent can summarize the current state of a feature request before you even open it. - Deep code context is now available to your AI client. The
get_file_contentstool lets you pull the exact text from any file, giving your agent enough data to answer complex questions about logic or structure. - Streamline bug tracking by opening tickets on demand. You can use
create_new_issueand follow up immediately withadd_issue_comment, all without logging into a separate ticketing system. - Search the code instantly, no GUI required. When you need to find a specific function or pattern across multiple files, rely on
search_github_codeinstead of manual keyword searches.
Real-World Use Cases
The PM needs project visibility.
A Project Manager has been asked for an update on a feature rollout. Instead of manually navigating to the repo's issues page, they ask their agent to 'list all open issues in vinkius/vurb-docs.' The MCP uses list_repo_issues and summarizes which tickets are stalled or waiting for review.
The Dev needs historical context.
A developer is debugging a tricky bug. Instead of searching through dozens of commits, they ask the agent to 'list commits in the main branch from last week.' The MCP uses list_commits and provides the commit hashes and summaries needed for immediate diagnosis.
The DevOps team needs an audit trail.
An engineer needs to verify that a specific piece of code used in a release is documented correctly. They ask the agent to 'get file contents for README.md' and instantly retrieve the source text, confirming compliance.
The Team Lead needs project coordination.
A team lead needs to report on all available code assets across departments. They prompt the agent to find 'all repositories belonging to our company organization,' using list_org_repositories for a complete inventory.
The Tradeoffs
Searching by vague keywords
Trying to ask your AI client, 'Tell me about the API structure.' This is too general and requires manual searching across multiple documentation pages.
→
Be specific. To find code examples, use search_github_code with a defined keyword or file name. If you need an overview of a project's scope, start by calling get_repository_details.
Mismanaging PR workflows
Trying to merge code without first knowing if the branch exists or if there are open conflicts, leading to failed merges.
→
Always check the status first. Use list_pull_requests to review all existing PRs and then use get_issue_details on any related issues before attempting to create a new pull request.
Forgetting connection checks
Assuming the agent knows your permissions or if your token expired, leading to generic 'Access Denied' errors.
→
Always start by verifying access. Use verify_api_connection first. This confirms that your authentication setup is current and functional before any heavy lifting.
When It Fits, When It Doesn't
Use this MCP if you need to manage the entire software development lifecycle—from tracking a bug (using create_new_issue) to analyzing the code fix (get_file_contents) and finally merging it (create_pull_request). It’s built for continuous, full-stack interaction with GitHub data.
Don't use this if you just need basic information that is available in a single, non-interactive view. For example, if you only want to know the name of a repository and nothing else, simply viewing the dashboard might be enough. If your task involves actions (creating, updating, commenting) or deep analysis across multiple files, this MCP handles it.
Common Questions About GitHub MCP
How do I list all my company's projects using the list_org_repositories tool? +
The agent uses list_org_repositories to gather a complete inventory of every repository under your organization. This is faster than navigating through individual team dashboards.
What does add_issue_comment do in the GitHub MCP? +
add_issue_comment allows you to reply directly to an existing issue or PR from your chat, adding context and updates without leaving the agent interface.
Can I find code snippets using search_github_code? +
Yes. search_github_code lets you run powerful searches across the entire repository's codebase to isolate specific functions or patterns, much faster than a simple keyword search.
How do I know if my API connection is working with verify_api_connection? +
Simply ask your agent to run verify_api_connection. The tool checks the authentication token and confirms that your MCP client has active, valid access rights to GitHub.
What if I need more info about a specific bug? Should I use get_issue_details? +
Yes. get_issue_details pulls all the deep metadata for one issue—the full conversation history, labels, and assignees—giving you the complete picture in one call.
If I want to track a feature or fix, how do I use the `create_pull_request` tool? +
You must specify both the head branch (your changes) and the base branch (the target merge location). This action initiates the formal pull request process so your agent can monitor its status and lifecycle.
When I use `get_file_contents` to read a file, what kind of data does it give me? +
It returns the raw text content of the specified file. You get pure code or documentation text, which you can then pass directly to your agent for analysis, summarization, or modification.
How do I know what my current permissions are on GitHub? Should I use `get_my_github_profile`? +
Yes. Calling get_my_github_profile provides detailed information about the authenticated user's identity and associated account context. Your agent uses this to confirm your access rights before attempting major actions.
Multi-server workflows that include GitHub MCP
Deploy Containers to Production Using MCP
Code pushed, images built, tags verified, deploys triggered, status reported , ship containers from commit to production in one prompt
Extract Architecture Principles Using MCP
Code patterns formalized, universal laws derived, causal forces identified , replace ad-hoc architecture with mathematical proof
Find Codebase Duplications Using MCP Servers
Your codebase has 4 different implementations of date formatting, 3 versions of the retry logic, and 2 competing validation libraries , but nobody knows because grep only finds exact matches and these duplicates are semantic
Generate Error Postmortems Automatically via MCP
Errors captured, stack traces analyzed, root cause commits identified, postmortem docs generated , write incident reports without the pain
How MCP Servers Auto-Triage Bug Reports
New bugs detected, severity classified, sprint tickets created, team notified , triage your backlog without a standup
MCP Recipe for Code Review Time Analytics
Review bottlenecks detected, unreviewed PRs surfaced, reviewer workload balanced, team velocity measured , fix your code review process with data
Use it with your favorite AI tools
Connect this server to Cursor, Claude, VS Code, and more.