# Harness Service Onboarding

> >

- **Type:** Skill
- **Install:** `agentstack add skill-dungnotnull-hybrid-harness-chaos-process-prm-s05-service-onboarding`
- **Verified:** Yes — security-reviewed for prompt injection and unsafe behavior
- **Seller:** [dungnotnull](https://agentstack.voostack.com/s/dungnotnull)
- **Installs:** 0
- **Category:** [Agent Skills](https://agentstack.voostack.com/c/agent-skills)
- **Latest version:** 0.1.0
- **License:** MIT
- **Upstream author:** [dungnotnull](https://github.com/dungnotnull)
- **Source:** https://github.com/dungnotnull/hybrid-harness-chaos-process-prm/tree/main/skills/s05-service-onboarding

## Install

```sh
agentstack add skill-dungnotnull-hybrid-harness-chaos-process-prm-s05-service-onboarding
```

Requires the [AgentStack CLI](https://agentstack.voostack.com/docs/cli). Works with Claude Code, Cursor, and any MCP-compatible agent.

## About

# Harness Service Onboarding

## Purpose
Guide the complete end-to-end onboarding of a new service onto Harness: from creating the service entity to wiring up environments, infrastructure definitions, connectors, monitored services, and the first pipeline run.

---

## Input Contract

| Input | Source | Required |
|---|---|---|
| Pipeline YAML / service name | s04 (workflow_context.artifacts) | Yes |
| Repository URL, container registry | s01 or user | Yes |
| Kubernetes cluster info | s06 output or user | Yes |
| Harness org/project identifiers | CLAUDE.md | Yes |
| Delegate selector | s06 output | No |
| Observability tool choice | s02 taste or s01 | No |

## Output Contract

| Output | Destination | Format |
|---|---|---|
| Service definition YAML | `.commandcode/artifacts/service-.yaml` | YAML |
| Environment YAML per tier | `.commandcode/artifacts/env-.yaml` | YAML |
| Infrastructure definition YAML | `.commandcode/artifacts/infra--.yaml` | YAML |
| Connector configs | `.commandcode/artifacts/connectors.yaml` | YAML |
| Onboarding hand-off document | User + s06 feed | Markdown |
| Service context (for s06, s07, s08, s10, s15) | workflow_context.artifacts | YAML object |

---

## Prerequisites
Gather before proceeding:
- [ ] Service name, language/runtime, and repository URL
- [ ] Container registry location (ECR, GCR, DockerHub, ACR)
- [ ] Kubernetes cluster and namespace (or other target infra)
- [ ] Harness Organization and Project identifiers
- [ ] Cloud provider credentials (for connector setup)
- [ ] Observability tool (Prometheus, Datadog, etc.) if CV is needed

---

## Onboarding Checklist (run in order)

```
[ ] 1. Create Connector(s)
[ ] 2. Create Service Definition
[ ] 3. Create Environment(s)
[ ] 4. Create Infrastructure Definition(s)
[ ] 5. Scope Delegate to Namespace
[ ] 6. Create Monitored Service (if CV enabled)
[ ] 7. Create Pipeline (reference harness/01-pipeline-design)
[ ] 8. Create Input Set for each environment
[ ] 9. First dry-run validation
[ ] 10. Hand-off documentation
```

---

## Step 1 — Connectors

### Kubernetes Cluster Connector
```yaml
connector:
  name:  K8s
  identifier: _k8s
  orgIdentifier: 
  projectIdentifier: 
  type: K8sCluster
  spec:
    credential:
      type: InheritFromDelegate
    delegateSelectors:
      - 
```

### Docker Registry Connector
```yaml
connector:
  name: 
  identifier: 
  type: DockerRegistry
  spec:
    dockerRegistryUrl: https://index.docker.io/v2/
    providerType: DockerHub
    auth:
      type: UsernamePassword
      spec:
        username: 
        passwordRef: docker_password_secret
```

### GitHub Source Connector
```yaml
connector:
  name: GitHub 
  identifier: github_
  type: Github
  spec:
    url: https://github.com/
    connectionType: Account
    authentication:
      type: Http
      spec:
        type: UsernameToken
        spec:
          username: 
          tokenRef: github_pat_secret
    apiAccess:
      type: Token
      spec:
        tokenRef: github_pat_secret
    delegateSelectors:
      - 
```

---

## Step 2 — Service Definition

```yaml
service:
  name: 
  identifier: 
  orgIdentifier: 
  projectIdentifier: 
  description: ""
  tags:
    team: 
    domain: 
    managed-by: hcprm
  serviceDefinition:
    type: Kubernetes
    spec:
      manifests:
        - manifest:
            identifier: k8s_manifests
            type: K8sManifest
            spec:
              store:
                type: Github
                spec:
                  connectorRef: github_
                  gitFetchType: Branch
                  branch: main
                  paths:
                    - k8s/
              valuesPaths:
                - k8s/values.yaml
              skipResourceVersioning: false
              enableDeclarativeRollback: true
      artifacts:
        primary:
          primaryArtifactRef: primary
          sources:
            - identifier: primary
              sourceType: DockerRegistry
              spec:
                connectorRef: 
                imagePath: 
                tag: 
      variables:
        - name: replicaCount
          type: String
          value: "2"
        - name: memoryLimit
          type: String
          value: "512Mi"
        - name: cpuLimit
          type: String
          value: "500m"
```

---

## Step 3 — Environments

```yaml
# Create one environment entity per tier
environment:
  name:   # e.g., staging
  identifier: 
  orgIdentifier: 
  projectIdentifier: 
  type: PreProduction  # or Production
  tags:
    managed-by: hcprm
  variables:
    - name: replicaCount
      type: String
      value: "1"   # override per-env
  overrides:
    manifests:
      - manifest:
          identifier: env_values
          type: Values
          spec:
            store:
              type: Github
              spec:
                connectorRef: github_
                gitFetchType: Branch
                branch: main
                paths:
                  - k8s/values-.yaml
```

---

## Step 4 — Infrastructure Definition

```yaml
infrastructureDefinition:
  name: --infra
  identifier: __infra
  orgIdentifier: 
  projectIdentifier: 
  environmentRef: 
  deploymentType: Kubernetes
  type: KubernetesDirect
  spec:
    connectorRef: _k8s
    namespace: 
    releaseName: release-
  allowSimultaneousDeployments: false
```

---

## Step 5 — Delegate Scoping

Ensure the delegate has access to the target namespace:

```yaml
# If using Harness Delegate Helm chart, apply RBAC:
apiVersion: rbac.authorization.k8s.io/v1
kind: Role
metadata:
  name: harness-delegate-role
  namespace: 
  labels:
    managed-by: hcprm
rules:
  - apiGroups: ["", "apps", "extensions", "batch"]
    resources: ["pods", "deployments", "services", "configmaps",
                "secrets", "replicasets", "jobs", "statefulsets"]
    verbs: ["get", "list", "watch", "create", "update", "patch", "delete"]
  - apiGroups: ["networking.k8s.io"]
    resources: ["ingresses"]
    verbs: ["get", "list", "watch", "create", "update", "patch"]
---
apiVersion: rbac.authorization.k8s.io/v1
kind: RoleBinding
metadata:
  name: harness-delegate-binding
  namespace: 
subjects:
  - kind: ServiceAccount
    name: harness-delegate
    namespace: harness-delegate
roleRef:
  kind: Role
  name: harness-delegate-role
  apiGroup: rbac.authorization.k8s.io
```

Reference `harness/03-delegate-management` for full delegate setup.

---

## Step 6 — Input Sets

Input sets capture environment-specific values, avoiding pipeline duplication.

```yaml
inputSet:
  name:  Input Set
  identifier: _input_set
  orgIdentifier: 
  projectIdentifier: 
  pipelineIdentifier: _pipeline
  inputSetReferences: []
  pipeline:
    identifier: _pipeline
    variables:
      - name: imageTag
        value: latest
      - name: targetEnv
        value: 
    stages:
      - stage:
          identifier: Deploy_
          spec:
            environment:
              environmentRef: 
              infrastructureDefinitions:
                - identifier: __infra
```

---

## Step 7 — Dry-Run Validation

Before first real deployment, run pipeline with `skipDryRun: false`:
1. Navigate to pipeline → Run → select input set
2. Enable **"Dry Run"** toggle (CD stage will generate manifests without applying)
3. Review generated manifest in execution logs
4. Confirm: correct image, namespace, resource limits, labels

---

## Onboarding Hand-off Template

Generate and share this document with the service team:

```markdown
## Harness Onboarding Summary — 

**Project**:  | **Org**: 

### Resources Created
| Resource | Identifier |
|---|---|
| Service |  |
| Environment (dev) | dev |
| Environment (staging) | staging |
| Infrastructure (dev) | _dev_infra |
| Infrastructure (staging) | _staging_infra |
| Pipeline | _pipeline |

### How to Deploy
1. Push to `main` branch → webhook trigger fires automatically
2. Or: Harness UI → Pipelines →  → Run → select input set

### Contacts
- SRE Owner: 
- Harness Admin: 
- Runbook: 
```

---

## AI Agent Integration

### Autonomy Level

| Aspect | Level | Description |
|---|---|---|
| Current | L2 | AI generates service/env/connector definitions from prompts |
| Target | L3 | AI creates and configures services, human validates |

### Harness AI Agent

**Agent**: Harness AI DevOps Agent
**Capabilities**:
- Service creation via conversational prompts
- Environment definition generation
- Connector configuration
- Service dependency mapping

### Human Gates

- Service definition approval
- Production environment creation

### Fallback

When Harness AI is unavailable: Use static pipeline templates from s09 Template Library and manual YAML construction following Harness schema documentation.

---

## Success Criteria
- [ ] All connectors tested and showing "Success"
- [ ] Service definition saved and visible in Harness UI
- [ ] Infrastructure definition linked to correct cluster/namespace
- [ ] Dry-run executed without errors
- [ ] First real deployment to dev completed successfully
- [ ] Hand-off document delivered to service team

## Source & license

This open-source skill is cataloged on AgentStack and links to its original source — we do not rehost the code.

- **Author:** [dungnotnull](https://github.com/dungnotnull)
- **Source:** [dungnotnull/hybrid-harness-chaos-process-prm](https://github.com/dungnotnull/hybrid-harness-chaos-process-prm)
- **License:** MIT

Install and usage instructions live in the source repository linked above.

## Pricing

- **Free** — Free

## Security capabilities

Automated source analysis of v0.1.0 — what this tool can access:

- **Network access:** no
- **Filesystem access:** no
- **Shell / process execution:** no
- **Environment & secrets:** yes
- **Dynamic code execution:** no

*"Yes" means the capability is present in the source — more access means more to trust, not that it is unsafe.*


## Versions

- **0.1.0** — security scan: passed — Imported from the upstream source.

## Links

- Listing page: https://agentstack.voostack.com/l/skill-dungnotnull-hybrid-harness-chaos-process-prm-s05-service-onboarding
- Seller: https://agentstack.voostack.com/s/dungnotnull
- Browse the marketplace: https://agentstack.voostack.com/browse

---
Listed on AgentStack — the marketplace for AI agent skills and MCP servers. Every listing is security-reviewed. Creators keep 70%.