Coveralls MCP for AI. Analyze code coverage and build health in conversation.
Gemini Works with every AI agent you already use
…and any MCP-compatible client
Connect to your AI in seconds.
Coveralls Code Coverage Analytics API tracks your code quality metrics by managing repositories, submitting coverage reports, and monitoring build statuses through natural conversation with your AI agent.
What your AI can do
Close parallel build
Closes an active, parallel build sequence within the Coveralls system.
Create repo
Creates a brand new repository entry on Coveralls for tracking coverage metrics.
Get build web data
Retrieves the raw JSON data from a specific build's web page, allowing deep analysis of its results.
Create or update the configuration details and tokens for multiple version control repositories.
Send detailed test reports, including source file data and Git metadata, to track coverage improvements.
Fetch structured JSON data about specific build web pages or entire repository health records for deep analysis.
Manage complex parallel builds by triggering final calculations or closing active build sequences.
Trigger a rerun of the build processing when you spot transient errors that need re-checking.
Ask an AI about this
Waiting for input…
Coveralls (Code Coverage Analytics API) 10 Tools
These tools let you perform every action required to track, submit, and manage code coverage data through the Coveralls system.
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 Coveralls (Code Coverage Analytics API) on VinkiusClose Parallel Build
Closes an active, parallel build sequence within the Coveralls system.
Create Repo
Creates a brand new repository entry on Coveralls for tracking coverage metrics.
Get Build Web Data
Retrieves the raw JSON data from a specific build's web page, allowing deep analysis...
Get File Web Data
Pulls the JSON representation for a single source file’s web page, useful for...
Get Job Web Data
Gets the raw JSON data from an individual job's web page, showing what happened...
Get Repo
Retrieves fundamental information about a repository from Coveralls using its API token.
Get Repo Web Data
Fetches the JSON data for an entire repository's web page, giving an overview of its history and status.
Rerun Build
Triggers a complete recalculation or rerun of existing build processing to resolve...
Submit Job
Submits a full coverage report (a job) using source files and Git metadata for...
Update Repo
Modifies the details of an existing repository on Coveralls, like changing its name...
Security and governance baked right in.
Pick your AI client below to get set up. Just create a Vinkius account, subscribe, and you're instantly up and running. We handle the entire backend infrastructure, delivering out-of-the-box support for HTTPS Streamable, SSE, and OAuth2—zero messy routing required.
Claude AI
Open Claude Settings
Go to claude.ai, click your profile icon, then navigate to Customize → Connectors.
Add Custom Connector
Click the "+" button and select Add custom connector. Paste your Vinkius endpoint URL:
https://edge.vinkius.com/[YOUR_TOKEN_HERE]/mcp
Replace [YOUR_TOKEN_HERE] with your token
from cloud.vinkius.com. For OAuth-protected servers, expand
Advanced settings to add credentials.
Start a conversation
Open a new chat. The Coveralls integration is available immediately — no restart needed.
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 Coveralls (Code Coverage Analytics API), then connect any of our 5,100+ other servers whenever your AI needs more. One click, no limits.
- Use this MCP plus 5,100+ 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 Coveralls. 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 connection provides 10 powerful capabilities that interface natively with Claude, ChatGPT, Cursor, and other compatible AI platforms. No middleware. No custom integration required.
Checking code quality used to mean clicking through half a dozen dashboards.
Today, checking coverage means logging into Coveralls, navigating to the repo, then finding the build number. You click on that, and another page loads. If you need details about a specific file or job from that build, you often have to copy key identifiers and paste them elsewhere just to get the final answer.
With this MCP, your agent does all the clicking for you. Instead of jumping between pages and manually retrieving data, you simply ask your client: 'What was the coverage for Job ID 123?' The agent handles the complex API calls—like calling `get_job_web_data`—and gives you the answer straight up.
Getting repository info with get_repo
Before starting any new project or analyzing a historical build, you have to know if the repo even exists and what its current tokens are. Manually checking this means digging through admin panels until you find the correct settings page.
Now, your agent uses `get_repo`. It pulls all the core repository details instantly. You get immediate confirmation of status and necessary tokens without ever logging into a dashboard.
What your AI can actually do with this
Your agent connects to this MCP when you need to know if a feature release is stable. Instead of navigating multiple CI/CD dashboards or digging through log files, you ask your AI client for the current code coverage status. It checks the latest reports and tells you exactly what's missing—which test suite failed or which file dropped below the acceptable quality threshold.
This MCP lets you manage everything from setting up new repositories to triggering a full build rerun if something looks suspicious. Because it provides deep visibility into your commit history and testing results, it’s essential for maintaining high code quality standards. Vinkius makes this connection easy; just link your agent once and gain access to all these developer tools.
019e5d0e-2ef6-720c-812e-9cb36870b353 
Here's how it actually works
The bottom line is, your AI client handles all the API complexity so you just get a clean status update and actionable next steps.
First, connect your agent and provide your Coveralls Personal Access Token.
Next, tell the agent what repository or job status you want to analyze (e.g., 'What's the coverage for my main branch?').
Finally, the agent calls the necessary tools to retrieve data or submit a report, giving you an immediate answer on code health.
Who is this actually for?
Anyone whose job involves making sure code actually works—DevOps Engineers, Software Developers, QA Leads. If checking build dashboards or submitting test reports is part of your routine, you need this.
Submitting coverage reports after local merges and running checks to ensure new features haven't broken existing tests.
Automating the setup of new project repositories or manually closing out complex, parallel CI/CD build chains.
Analyzing coverage regressions across branches and managing build thresholds to keep code reliability high.
What Changes When You Connect
Stop switching between tabs. Instead of checking a dashboard, you ask your agent to get the repository info using get_repo and instantly know if the current branch meets quality standards.
Save time debugging pipelines. If a build seems stuck or incomplete, use close_parallel_build to clean up the state and prevent resource leaks without manual intervention.
Improve report accuracy. When you suspect an issue, don't just rely on visual checks; run get_job_web_data to pull the raw JSON data for programmatic analysis.
Handle regressions faster. If coverage drops, use submit_job immediately from your terminal with fresh test results rather than waiting for a full pipeline cycle.
Maintain stable systems. Use rerun_build when you suspect a build failed due to temporary network issues, letting the agent handle the retry logic.
See it in action
Investigating a Sudden Drop in Coverage
A developer notices coverage dropped below 90% after a merge. They ask their agent to analyze the last build's raw data, triggering get_build_web_data. The agent identifies which source file (get_file_web_data) is missing test cases so they can fix it immediately.
Setting up a New Microservice
The DevOps engineer needs to track a new service. They first use create_repo to initialize the coverage tracking, then connect their CI/CD pipeline to submit initial reports using submit_job.
Dealing with Stalled Builds
The QA lead sees multiple parallel builds running that are no longer needed. They ask the agent to manage them, which calls close_parallel_build to clean up the build environment and keep resources clear.
Troubleshooting Build Failures
A critical feature build failed inexplicably. Instead of restarting the entire pipeline manually, they ask the agent to run rerun_build. The agent triggers the rerun and monitors the result until success.
The honest tradeoffs
Treating status data as final.
A developer assumes the visible build number is accurate because they just saw it on the dashboard. They try to use that number for everything, wasting time and getting incorrect results.
To get reliable data, always start by using get_repo or get_repo_web_data to confirm the repository's current state before attempting any reads or writes.
Overwriting settings manually.
An engineer tries to update a repo token by simply changing it in the web UI, but they aren't sure if all associated build rules will follow.
To guarantee an accurate change, use update_repo through your agent. This ensures the new configuration is applied programmatically across all connected systems.
Missing context in reports.
A developer runs a test and only submits the files without necessary metadata, resulting in an incomplete or unusable coverage report.
When submitting a job, always provide full Git metadata alongside your source files. Use submit_job to send complete packets of data so the reports are usable.
When It Fits, When It Doesn't
Use this MCP if your primary need is deep visibility into code quality and build execution status. You must be tracking what percentage of your code base was tested, or you need to manage the lifecycle (creating, updating, submitting reports) of those test results. Don't use this if you just need to send a simple message or check general operational metrics; for that, look at messaging MCPs. If you only want to read static data without changing anything, get_repo and get_build_web_data are your best tools. But remember: if the repository structure itself is wrong, start with update_repo; never assume state.
Questions you might have
How do I submit coverage reports using submit_job? +
You must provide the service name, job ID, source files, and Git metadata. The agent uses submit_job to package all this data correctly for Coveralls.
What's the difference between get_repo and get_repo_web_data? +
get_repo gives you high-level, fundamental information about the repository. get_repo_web_data pulls a much larger chunk of JSON data from the web page, giving deeper historical insights.
How can I fix a build that fails due to temporary issues? +
You use rerun_build. This tells your agent to trigger a full re-process of the failed build. The agent handles talking to Coveralls to make sure the recalculation starts.
Can I track multiple repositories with this MCP? +
Yes, you can manage several repos by first calling get_repo for each one and then using those credentials to run other tools like update_repo across the board.
What is the difference between using `get_repo` and fetching web data with `get_repo_web_data`? +
The API call gives you structured JSON objects, perfect for direct processing. Web data retrieval fetches a raw JSON representation of what's visible on the actual repository page, which is useful if you need to analyze visual trends or complex layouts not exposed in the core metadata.
How do I manage multiple concurrent build processes using `close_parallel_build`? +
close_parallel_build officially terminates a set of parallel builds. This function is essential for managing CI/CD workflows where multiple jobs run simultaneously; it ensures the entire build sequence completes or fails cleanly, preventing orphaned processes.
If I encounter an API rate limit when using `get_job_web_data`, how do I handle it? +
Rate limits mean you're making too many requests in a short window. Wait about one minute and try the call again. For continuous monitoring, consider batching your data calls or implementing exponential backoff logic in your agent code.
Does `create_repo` require specific permissions beyond the personal API token? +
Yes, the provided personal API token must have write access to create new repository records. If you get a permission error, check that your token scope includes resource management rights for Coveralls.
How do I retrieve my repository's secret token for CI configuration? +
Use the get_repo tool by providing the service (e.g., 'github') and repository name. The agent will return the repository details, including the repo_token needed for your CI environment variables.
Can I submit a coverage report for a specific CI job manually? +
Yes! The submit_job tool allows you to send coverage data directly. You'll need to provide the repo_token, service_name, service_job_id, and the source_files JSON array containing coverage metrics.
How do I finalize a parallel build once all individual jobs are finished? +
Use the close_parallel_build tool. Provide your repo_token and the build_num. This triggers Coveralls to aggregate all parallel jobs and calculate the final coverage percentage for the build.
We've already built the connector for Coveralls. Just plug in your AI agents and start using Vinkius.
No hosting. No infrastructure. No complex setup.
All 10 tools are live and waiting.
You're up and running in seconds.
Gemini Vinkius gives your AI agents access to the full catalog of app connectors, all fully managed, secure, and enterprise-ready. One subscription, every tool you need.
Built, hosted, and secured by Vinkius. You just connect and go.