Create Agent Template

What it does

When your agent uses it

• Asked to create an agent template
• Build an AI agent
• Create an automation agent
• Generate agent pipeline files

Example prompts

• “/create-agent-template”

Requirements

• A credential in API_KEY
• Compatibility (from SKILL.md): No MCP server required. Generates YAML files locally.

Workflow steps

4 steps, taken from the first numbered list in SKILL.md.

1. 1Identify the agent's purpose - What task should the agent automate (code review, security scanning, test generation, documentation)?
2. 2Define inputs and configuration - What parameters does the agent need (repo, branch, connector, secrets)?
3. 3Generate three files - metadata.json (template metadata), pipeline.yaml (v1 syntax), and wiki.MD (user documentation)
4. 4Validate consistency - Ensure input references in pipeline.yaml match definitions in metadata.json

What it can do on your machine

Read from SKILL.md and the folder at commit c25faee. It shows what the files ask for, not the result of running them.

• Tool permissions

  Pre-approves nothing: there is no allowed-tools line, so your agent's usual permission prompts apply.

  From allowed-tools in the SKILL.md frontmatter.

• Runs code

  Shell commands in SKILL.md call:

  • git

  From the folder's file list and the shell code blocks in SKILL.md.

• Network

  No URLs in SKILL.md. Its commands use git, which can reach the network depending on how they are called.

  From URLs in SKILL.md, links to its own repository left out.

• Credentials

  Names these keys or tokens, usually read from environment variables:

  • API_KEY

  From names ending in _API_KEY, _TOKEN, _SECRET, _KEY or _PASSWORD in SKILL.md.

• Compatibility

  No MCP server required. Generates YAML files locally.

  From compatibility in the SKILL.md frontmatter.

Context cost

Create Agent Template loads about 2.2k tokens when it runs, and up to ~4.8k if it reads all its reference files. Until then it costs about 149 tokens; SKILL.md has 522 words of instructions outside code blocks.

Always · name and description, kept in context so the agent knows when to use it

~149

When it runs · the whole SKILL.md, loaded when a task matches

~2.2k

With references · SKILL.md plus every file in references/, read only if the agent opens them

~4.8k

Estimates: characters ÷ 4, the usual rule of thumb; real counts depend on the model's tokenizer. Scripts and assets cost tokens only if the agent reads them.

Safety

Automated static check — not a guarantee. Review scripts before installing. It scans the text of SKILL.md for risky patterns (piping downloads into a shell, reading credential files, hidden Unicode, destructive commands); files beside SKILL.md are not scanned.

SKILL.md

The full file from harness/harness-skills at commit c25faee, republished under its Apache-2.0 licence (© harness). 522 words, ~2,167 tokens.

name

create-agent-template

description

Generate Harness Agent Template files for AI-powered automation agents. Produces metadata.json, pipeline.yaml (v1 syntax), and wiki.MD files. Agents automate tasks like code review, security scanning, test generation, and documentation. This is a YAML generation skill. Use when asked to create an agent template, build an AI agent, create an automation agent, generate agent pipeline files, or set up a Harness agent. Trigger phrases: create agent template, agent template, AI agent, automation agent, build agent, code review agent, security scanner agent, Harness agent.

compatibility

No MCP server required. Generates YAML files locally.

metadata.author

Harness

metadata.mcp-server

harness-mcp-v2

metadata.version

1.0.0

license

Apache-2.0

templates/<agent-name>/
  metadata.json
  pipeline.yaml
  wiki.MD

{
  "name": "Agent Name",
  "description": "Brief description of what this agent does (1-2 sentences)",
  "version": "1.0.0"
}

version: 1
pipeline:
  clone:
    depth: 1
    ref:
      name: <+inputs.branch>
      type: branch
    repo: <+inputs.repo>
    connector: "<+inputs.gitConnector != null ? inputs.gitConnector.id : ''>"

  stages:
    - name: <stage-name>
      steps:
        - name: <step-name>
          run:
            container:
              image: <registry>/<image>:<tag>
            with:
              param: value
            env:
              SECRET_VAR: <+inputs.secretInput>
      platform:
        os: linux
        arch: arm64

  inputs:
    repo:
      type: string
      required: true
    branch:
      type: string
      default: main
    llmConnector:
      type: connector
    gitConnector:
      type: connector

clone:
  depth: 1
  ref:
    name: <+inputs.branch>
    type: branch
  repo: <+inputs.repo>
  connector: "<+inputs.gitConnector != null ? inputs.gitConnector.id : ''>"

clone:
  depth: 1000
  ref:
    type: pull-request
    number: <+inputs.pullReq>
  repo: <+inputs.repo>

inputs:
  repo:
    type: string
    required: true
  branch:
    type: string
    default: main
    description: Branch to analyze
  apiKey:
    type: secret
    default: account.my_secret
  llmConnector:
    type: connector
    description: LLM provider connector
  gitConnector:
    type: connector
    description: Git repository connector

- name: my-step
  run:
    container:
      image: registry/image:tag
    with:
      plugin_param: value
    env:
      API_KEY: <+inputs.apiKey>

- name: shell-step
  run:
    shell: bash
    script: |-
      echo "Running script"
      git add -A
      git diff --cached
    env:
      MY_VAR: value

# Input references
<+inputs.variableName>

# Connector token
<+inputs.connectorName.token>

# Step outputs
<+pipeline.stages.STAGE.steps.STEP.output.outputVariables.VAR>

# Environment variables
<+env.HARNESS_ACCOUNT_ID>
<+env.HARNESS_ORG_ID>
<+env.HARNESS_PROJECT_ID>

# Alternative syntax for inputs in env blocks
${{inputs.repo}}

# Agent Name

**Version:** 1.0.0
**Name:** Agent Name

## Overview

What this agent does and why it is useful.

---

## Key Capabilities

- **Capability 1**: Description
- **Capability 2**: Description

---

## How It Works

1. **Step 1**: Description
2. **Step 2**: Description
3. **Step 3**: Description

## Required Inputs

| Input | Type | Description | Default |
|-------|------|-------------|---------|
| `repo` | string | Repository identifier | -- |
| `branch` | string | Branch to analyze | main |

## Usage Example

\`\`\`yaml
inputs:
  repo: "my-org/my-repo"
  branch: "main"
\`\`\`

## Troubleshooting

### Common Issue 1
Solution description.

/create-agent-template
Create an agent template for code review that analyzes pull requests,
provides feedback, and posts review comments

/create-agent-template
Generate a security scanner agent that finds vulnerabilities and
creates a report

/create-agent-template
Create an agent that analyzes a codebase and generates documentation,
then opens a PR with the results

/create-agent-template
Build an agent template that generates unit tests to improve code coverage

© harness, Apache-2.0. Rendered from Markdown: HTML in the file is shown as text, images as links, and headings moved down two levels. Raw file

Files

SKILL.md and 1 other file (references) in skills/create-agent-template of harness/harness-skills.

• SKILL.md
• references/agent-examples.md

Open the folder on GitHubat commit c25faee