infra-jenkins-pipeline-ops

Jenkins Pipeline Operations

Purpose

Use this Skill to inspect or operate a Jenkins instance, review a Jenkinsfile, or diagnose a Jenkins Pipeline. It works with self-hosted and managed Jenkins deployments and does not assume a particular network, container host, registry or source-control server.

Required Configuration

Configure the companion CLI in ~/.harness-ai-kit/config.yaml:

assets:
  jenkinsctl:
    jenkins_url: https://jenkins.example.com/
    jenkins_user: automation-user
    jenkins_api_token: <store-outside-source-control>

The checked-in config.defaults.yaml defines the complete schema. Command-line values override the global configuration. Environment variables are CI-only fallbacks; do not put tokens in a repository or an agent instruction file.

Workflow

1. Confirm the target Jenkins URL, intended action and whether it changes shared platform state.
2. Run jenkinsctl doctor and use read-only job/configuration commands first.
3. Export the current job configuration before an update, disable or delete.
4. Keep credentials in the Jenkins credential store and reference their IDs; never put credential material in a Jenkinsfile.
5. After a write, read the relevant job/configuration back and report the effective result.

Guardrails

• Prefer the REST API or jenkins-cli.jar; use SSH CLI only when explicitly configured by the operator.
• Treat job deletion, credential changes, plugin changes and global tool or shared-library configuration as high-impact operations.
• jenkinsctl applies role checks from ~/.harness-ai-kit/config.yaml. Read-only commands are available to consumers; daily delivery actions need contributor access; shared platform and credential changes need maintainer access.
• Use generic service URLs and documentation IP ranges in examples. Do not copy hostnames, private addresses, tokens, credentials or deployment topology into public material.