- Home
- Runbook Automation Software
- RunbookAI
RunbookAI
What it is
RunbookAI is a free, open-source incident-response tool for SRE teams handling production workloads on AWS and Kubernetes. It investigates incidents by collecting context, ranking possible explanations, checking those hypotheses against infrastructure, identifying a root cause, and suggesting remediation. Natural-language queries cover AWS, Kubernetes, and CloudWatch. The tool can execute step-by-step runbooks, while infrastructure queries remain read-only by default and changes require explicit human approval. RunbookAI indexes knowledge from local files, Confluence, and Google Drive. Documented integrations include AWS, Kubernetes, PagerDuty, OpsGenie, Slack, and Claude Code. Its self-hosted shared knowledge server exposes runbooks, postmortems, and known issues through REST and an MCP endpoint. CLI mode is also available; there is no hosted service or control plane. It runs in the user’s infrastructure, with no telemetry according to the maker, and sends LLM calls directly to the configured provider. Local audit logs record hypotheses, evidence, approvals, and execution results.
For and against
For
- Open-source MIT License
- Read-only queries by default
- Human approval for mutating actions
- Local audit trail
- Shared knowledge server
Against
- Linux platform only
- Requires Bun or Node.js
- Requires an Anthropic API key
- No hosted service or control plane
The verdict
RunbookAI combines hypothesis-driven incident investigation with controlled runbook execution and shared operational knowledge. Its self-hosted design, approval gates, and local logging fit teams that need infrastructure-focused response controls.
Who it suits
RunbookAI suits SRE teams operating production systems on AWS and Kubernetes. It is intended for teams wanting self-hosted incident investigation, runbooks, and shared operational knowledge.
On the record
- Audit trail
- RunbookAI logs investigation hypotheses, evidence, proposed actions, approval decisions and execution results locally.userunbook.ai · read 4 Oct 2026
- Claude Code
- The shared knowledge server exposes an MCP endpoint that Claude Code can use to query operational knowledge.userunbook.ai · read 4 Oct 2026
- Data handling
- The maker says RunbookAI runs within the user's infrastructure, sends no telemetry, and sends LLM calls directly to the configured provider without proxying or storing prompts.userunbook.ai · read 4 Oct 2026
- Deployment
- RunbookAI offers CLI mode and a self-hosted shared knowledge server, with no hosted service or control plane.userunbook.ai · read 4 Oct 2026
- Deployment and license
- RunbookAI is self-hosted and released under the MIT License.userunbook.ai · read 4 Oct 2026
- Incident workflow
- Its investigation workflow gathers incident context, ranks hypotheses, tests them against infrastructure, identifies a root cause and suggests remediation.userunbook.ai · read 4 Oct 2026
- Infrastructure queries
- The CLI supports natural-language queries across AWS, Kubernetes, and a knowledge base.userunbook.ai · read 4 Oct 2026
- Integrations
- Documented integrations include AWS, Kubernetes, PagerDuty, OpsGenie, Slack and Claude Code.userunbook.ai · read 4 Oct 2026
- Intended users
- The maker describes it as an incident-response tool for SRE teams and is onboarding teams running production workloads on AWS and Kubernetes.userunbook.ai · read 4 Oct 2026
- Investigation
- Its investigation workflow gathers incident context, tests ranked hypotheses against infrastructure, and suggests remediation steps.userunbook.ai · read 4 Oct 2026
- Knowledge sources
- RunbookAI can index knowledge from local files, Confluence and Google Drive.userunbook.ai · read 4 Oct 2026
- Open source
- RunbookAI is released under the MIT License.userunbook.ai · read 4 Oct 2026
- Purpose
- RunbookAI investigates production incidents by forming hypotheses, gathering evidence and recommending fixes.userunbook.ai · read 4 Oct 2026
- Requirements
- The documentation lists Bun 1.0+ or Node.js 20+ and an Anthropic API key for Claude as requirements.userunbook.ai · read 4 Oct 2026
- Runbooks
- It can execute step-by-step runbooks, with approval gates for mutating actions.userunbook.ai · read 4 Oct 2026
- Safety
- Infrastructure queries are read-only by default, and every mutating action requires explicit human approval.userunbook.ai · read 4 Oct 2026
- Safety controls
- Every mutating action requires explicit human approval, while infrastructure queries are read-only by default.userunbook.ai · read 4 Oct 2026
- Shared knowledge server
- A self-hosted server lets a team query shared runbooks, postmortems and known issues through a REST API and an MCP endpoint.userunbook.ai · read 4 Oct 2026
