Developers are increasingly working inside AI-powered coding tools like Claude Code, Cursor, and Codex.

The workflow is changing.

Instead of constantly switching between dashboards, logs, terminals, and configuration files, developers are starting to collaborate directly with coding agents using natural language.

We think CI/CD should evolve alongside that shift.

Today, we’re introducing Semaphore for AI Agents — a new open-source CLI and agentic interface designed to make Semaphore fully accessible from AI coding agents.

This is the first step toward what we call the AI-native Semaphore experience.

What is Semaphore for AI Agents?

Semaphore for AI Agents gives coding assistants a structured way to interact with Semaphore.

Instead of manually navigating CI/CD systems, developers can simply ask:

* “Why is my CI failing?”

* “What tests are flaky?”

* “Show me the critical path in this pipeline.”

* “Summarize the health of this project.”

And their coding agent can retrieve, analyze, and act on that information directly through Semaphore.

The first release includes:

* Pipeline diagnostics

* Flaky test detection

* Critical path analysis

* Organization-wide CI/CD insights

* MCP support

* Claude Code integrations

* Remote execution workflows on Semaphore infrastructure

We also demonstrate how developers can provision ephemeral machines for agent-driven workflows, remote testing, and scalable execution.

Built for Agentic Workflows

Semaphore for AI Agents was designed specifically for AI-native development workflows.

The project ships with:

* Agent-oriented commands

* Structured JSON outputs

* Claude Code skills

* Generic agent skills

* A local MCP server

This allows coding assistants to interact directly with Semaphore while developers stay inside their coding environment.

Fully Open Source

Semaphore for AI Agents is fully open source.

Developers can inspect how it works, extend workflows, contribute new commands, and build their own automations.

We believe AI-powered developer tooling should remain transparent, inspectable, and developer-controlled.

This Is Just the Beginning

Semaphore for AI Agents is the foundation for a broader direction we’re building toward at Semaphore:

* Developers define intent

* Agents handle repetitive execution

* Semaphore provides the infrastructure and orchestration layer

Over the coming weeks, we’ll continue shipping new workflows, MCP capabilities, testing automation, and scalable agent execution features.

Watch the Demo

We recorded a full walkthrough showing:

* CI/CD debugging workflows

* MCP integrations

* Claude Code usage

* Organization-wide insights

* Remote execution on Semaphore infrastructure

→ [Read the full blog post]

Thanks for following along.

Till the next product update.

Pete Miloravachttps://semaphore.io

This is a public episode. If you would like to discuss this with other subscribers or get access to bonus episodes, visit semaphoreio.substack.com