groundcover, the world’s leading bring-your-own-cloud (BYOC), eBPF and OpenTelemetry (OTel)-native observability platform, today announced a multi-surface launch that extends its observability intelligence to Agent Mode, external AI coding agents through the Model Context Protocol (MCP), and Slack. The launch gives engineers and their AI agents production context wherever they work, letting external agents either query groundcover’s telemetry directly or delegate open-ended investigations to the groundcover Agent.

AI coding agents can already read a codebase, review a pull request and reason about how software should behave, but they generally can’t explain what happened after that code reached production. groundcover addresses this by giving agents such as Claude, Cursor, Kiro and Codex access to logs, metrics, traces and Kubernetes events through direct MCP tools, plus a way to hand off deeper investigations, including root cause analysis, dashboard generation and monitor generation, to groundcover’s own observability agent.

“Coding agents can tell you what should happen, but they can’t tell you what already did happen in production unless they have access to the evidence,” said Shahar Azulay, CEO and co-founder of groundcover. “This launch means an agent doesn’t have to become an observability expert to get that evidence. It can query groundcover directly or delegate the investigation and keep working while groundcover does the digging.”

The launch supports two ways for an external agent to work with groundcover:

  • Direct access. When an agent already knows what evidence it needs, it can query groundcover’s MCP tools directly for logs, traces, metrics, Kubernetes events, live entities, monitors and related metadata.

  • Delegated investigation. When an agent knows there’s a problem but not where to look, it can hand the task to the groundcover Agent through asynchronous Agent Tasks. groundcover returns a job ID and continues the investigation in the background while the calling agent keeps working, then returns findings such as a completed root cause analysis or a draft dashboard.

Production Context Across Three Surfaces

groundcover’s intelligence now powers three entry points:

  1. Agent Mode: groundcover’s native environment for investigating production, generating dashboards and monitors, and steering an investigation in real time.

  2. IDEs and external agents: Connecting tools such as Kiro, Cursor, Claude and Codex to groundcover’s production context through MCP.

  3. Slack: @groundcover can join an incident thread, use the conversation as context, and pull in connected engineering systems such as Linear and GitHub.

Each surface runs on groundcover’s BYOC architecture, keeping telemetry inside the customer’s own cloud. A single eBPF sensor captures application and infrastructure behavior without requiring teams to instrument every service, and groundcover’s node-based pricing means customers aren’t penalized for retaining more data at higher fidelity, which is the evidence an agent needs to investigate accurately.

The multi-surface launch is available now. To learn more about agent-to-agent observability and how to connect groundcover to your IDE, agent or Slack workspace, visit https://www.groundcover.com/product/agent-mode.

About groundcover

groundcover is the full-stack observability platform for engineers & agents, allowing teams to monitor everything they run in their cloud, without compromising on data, control, or cost. It’s deployed inside the customer’s cloud premises and provides complete visibility into applications, infrastructure, and AI workloads without operational overhead. The platform offers unlimited data coverage at a fraction of the cost of legacy observability tools. Founded in 2021, the company is backed by leading global investors, including One Peak, Morgan Stanley Expansion Capital, Zeev Ventures, Angular Ventures, Heavybit and Jibe. Learn more at https://www.groundcover.com.

Media gallery

About The Author
Business Wire