create-service
Generate Harness Service YAML for deployable workloads and create via MCP. Supports Kubernetes, Helm, ECS, Serverless, SSH, and WinRm deployment types with artifact sources from Docker Hub, ECR, GCR, ACR, Nexus, and S3. Use when asked to create a service, define a Kubernetes service, set up a Helm chart deployment, configure an ECS service, or define what gets deployed. Trigger phrases: create service, service definition, Kubernetes service, Helm service, ECS service, deployment service, artifact source.
68
82%
Does it follow best practices?
Impact
—
No eval scenarios have been run
Passed
No known issues
Quality
Discovery
100%Based on the skill's description, can an agent find and select it at the right time? Clear, specific descriptions lead to better discovery.
This is an excellent skill description that clearly articulates specific capabilities (Harness Service YAML generation with multiple deployment types and artifact sources), provides explicit trigger guidance with a 'Use when' clause and dedicated trigger phrases, and is highly distinctive due to its Harness-specific context. The description is comprehensive yet concise, using proper third-person voice throughout.
| Dimension | Reasoning | Score |
|---|---|---|
Specificity | Lists multiple specific concrete actions: generating Harness Service YAML, creating via MCP, and enumerates specific deployment types (Kubernetes, Helm, ECS, Serverless, SSH, WinRm) and artifact sources (Docker Hub, ECR, GCR, ACR, Nexus, S3). | 3 / 3 |
Completeness | Clearly answers both 'what' (generate Harness Service YAML for deployable workloads with specific deployment types and artifact sources) and 'when' (explicit 'Use when...' clause with multiple trigger scenarios and a dedicated trigger phrases list). | 3 / 3 |
Trigger Term Quality | Includes a rich set of natural trigger phrases users would say: 'create service', 'service definition', 'Kubernetes service', 'Helm service', 'ECS service', 'deployment service', 'artifact source', plus contextual phrases like 'define what gets deployed' and 'set up a Helm chart deployment'. | 3 / 3 |
Distinctiveness Conflict Risk | Highly distinctive due to the specific Harness platform context, MCP integration, and the enumeration of specific deployment types and artifact sources. Unlikely to conflict with generic service or deployment skills. | 3 / 3 |
Total | 12 / 12 Passed |
Implementation
64%Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.
This is a solid, actionable skill with comprehensive YAML examples covering all major deployment types and artifact sources. Its main weaknesses are the lack of explicit validation/verification steps integrated into the workflow (connector existence checks, YAML validation before MCP creation) and the somewhat monolithic structure that inlines all deployment type examples rather than referencing them from a separate file. The content is well-organized but could be more concise by reducing redundancy between the base structure and the Kubernetes example.
Suggestions
Integrate validation steps directly into the workflow: add a step between 'Generate YAML' and 'Create via MCP' that explicitly checks connector existence using harness_list, and include a retry loop for creation failures.
Move the deployment type YAML examples into a separate reference file (e.g., `deployment-types.md`) and keep only the base structure and one representative example in SKILL.md to reduce length.
Remove the redundancy between the 'Service Structure' base example and the 'Kubernetes Service' example, which are nearly identical.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The skill is fairly efficient but includes some redundancy—the full service structure example overlaps significantly with the Kubernetes deployment type example. The artifact source types list and deployment type examples are useful reference material but could be more compact. The 'Performance Notes' section contains somewhat generic advice. | 2 / 3 |
Actionability | The skill provides fully executable YAML examples for each deployment type, concrete MCP tool calls with specific parameters, clear detection rules for auto-selecting service types, and specific artifact source configurations. All examples are copy-paste ready with real field names and values. | 3 / 3 |
Workflow Clarity | The 5-step workflow (detect → detect artifact → confirm → generate → create) is clearly sequenced and includes a user confirmation step. However, there's no explicit validation checkpoint before creating via MCP (e.g., validate YAML structure, verify connectors exist is mentioned in Performance Notes but not integrated into the workflow steps), and no error recovery loop. | 2 / 3 |
Progressive Disclosure | The content references `create-pipeline/references/codebase-analysis.md` for the full decision tree, which is good progressive disclosure. However, the skill itself is quite long (~180 lines of YAML examples) and could benefit from splitting deployment type examples into a separate reference file. No bundle files are provided to verify the referenced path exists. | 2 / 3 |
Total | 9 / 12 Passed |
Validation
100%Checks the skill against the spec for correct structure and formatting. All validation checks must pass before discovery and implementation can be scored.
Validation — 11 / 11 Passed
Validation for skill structure
No warnings or errors.
- Repository
- harness/harness-ai
- Commit
a000afa
- Reviewed
Table of Contents
Is this your skill?
If you maintain this skill, you can claim it as your own. Once claimed, you can manage eval scenarios, bundle related skills, attach documentation or rules, and ensure cross-agent compatibility.