# Cloudbees Oc

> CloudBees Operations Center integration for cross-controller deployment visibility and feature flag correlation during RCA

- Skill: `arvo-ai/cloudbees-oc` (Agent Skill)
- Install (CLI): `npx skillmds@latest add arvo-ai/cloudbees-oc`
- Raw SKILL.md: https://api.skillmd.com/api/skills/arvo-ai/cloudbees-oc/raw
- Safety review: pending
- Works with: Claude Code, Claude.ai, OpenAI Codex
- Category: AI & ML
- Author: Arvo-AI (https://skillmd.com/u/arvo-ai)
- Updated: 2026-09-22
- Page: https://skillmd.com/skills/arvo-ai/cloudbees-oc

---


# CloudBees Operations Center Integration

## Overview
CloudBees Operations Center (CJOC) integration for enterprise-wide CI/CD visibility during Root Cause Analysis.
Operations Center manages multiple Jenkins controllers and provides a unified view of all builds and deployments across the organization.

## Instructions

### Tool: cloudbees_rca

The same `cloudbees_rca` tool is used, with additional enterprise actions enabled when Operations Center is connected.

**Enterprise Actions:**
- `controller_list` — List all managed Jenkins controllers and their status (online, offline, failing)
- `cross_controller_deployments` — Query recent builds across ALL managed controllers; optional `service` filter and `time_window_hours`
- `flag_changes` — Query recent feature flag changes from CloudBees Feature Management; optional `app_id` filter

### RCA Investigation Flow

When Operations Center is connected, start broad and narrow down:

1. `cloudbees_rca(action='controller_list')` — See all managed controllers and their health
2. `cloudbees_rca(action='cross_controller_deployments', service='SERVICE')` — Find recent deployments across ALL controllers
3. `cloudbees_rca(action='flag_changes')` — Check if any feature flags were toggled near the incident time
4. Then drill into a specific build using `build_detail`, `pipeline_stages`, `build_logs` as needed

### When to Use

- **Deployment-related incidents**: Use `cross_controller_deployments` to see all recent activity across the entire CI/CD infrastructure, not just one controller.
- **Feature flag incidents**: Use `flag_changes` when an incident might be caused by a feature flag toggle (gradual rollout gone wrong, kill switch activated, etc.).
- **Controller health**: Use `controller_list` to check if a controller itself is unhealthy (offline, failing builds).

### Important Rules
- Always start with `cross_controller_deployments` for broad visibility before drilling into a specific controller.
- Use `flag_changes` when the incident pattern suggests a gradual rollout or toggle (e.g., affecting a percentage of users).
- Controller URLs returned by `controller_list` can be used with standard CloudBees CI actions (`build_detail`, etc.).