agentsclimarketplace

02 design agent

Skill NamhaeSusan/build-my-agent/skills/02-design-agent

Design an operational agent based on codebase analysis, determining tools, config, and system prompt contentFrom its SKILL.md

Install
npx -y skills add NamhaeSusan/build-my-agent --skill 02-design-agent

Assembled from the repository path, not quoted from the project. Check it against their README if it does not work.

2 things to look at

  • no licenseNo license file was found in the repository. Code published without one is not open source by default, so using it at work is a question for whoever answers licensing questions where you are.
  • 2 stars2 stars. Stars are a popularity signal and not a quality one, but at this level it is likely that nobody has read this closely except its author, and you would be relying on your own review.

SKILL.md

2.8 KB, as published. Nobody here has run it

Design Operational Agent

You are designing an operational agent based on the codebase analysis from the previous step.

Prerequisites

  • The <component>-analysis.md file exists from the analyze step
  • The component owner is available for Q&A

Process

Step 1: Load Analysis

Read the <component>-analysis.md file. Summarize the key findings to the owner:

"Based on the analysis, here's what I'm planning for your ops agent:

  • Tools: [tool list]
  • Key monitoring areas: [what the agent will watch]
  • Domain knowledge: [what the agent will know about your component]"

Step 2: Determine Tool List

Always included:

  • log_search — OpenSearch log querying
  • metric_query — Prometheus metric querying

Add based on analysis:

  • If Kafka dependencies found → suggest kafka_consumer tool (check lag, read messages)
  • If database found → suggest db_query tool (health check, slow query analysis)
  • If external APIs found → suggest api_health tool (endpoint health checks)

For each additional tool, ask:

"I'm suggesting a <tool_name> tool for <reason>. Should I include it? If yes, what endpoint/credentials does it need?"

Step 3: Define Config Values

Compile all config values needed for agent.yaml:

component: <from analysis>
description: <from analysis>

opensearch:
  endpoint: <ask if not in analysis>
  index_pattern: <from analysis>
  verify_certs: true

prometheus:
  endpoint: <ask if not in analysis>
  namespace: <from analysis>

Present to owner for confirmation.

Step 4: Draft System Prompt Content

Based on the analysis, prepare:

  1. Component overview — what it does, in operational context
  2. Domain knowledge — key error codes, business logic the agent should understand
  3. Failure patterns — from analysis, formatted as agent instructions
  4. Monitoring guidelines — what to check first, correlation patterns

Ask the owner:

"Here's the domain knowledge I'll give the agent: [draft content]

Is there anything to add or correct?"

Step 5: Present Design Summary

Present the complete design:

Agent Design for <component_name>
===================================

Tools:
  - log_search (OpenSearch)
  - metric_query (Prometheus)
  - <additional tools>

Config:
  <agent.yaml content>

System Prompt Sections:
  - Component: <summary>
  - Domain Knowledge: <summary>
  - Failure Patterns: <count> patterns documented
  - Guidelines: <summary>

"Does this design look good? If yes, I'll generate the agent project. Run: 03-generate-agent"

Output

Save the design as <component>-design.md with all the above details, structured so the generation skill can reference it directly.

Keep looking

Skills are one crate of 609,689. Ordering is by how many stacks a row turns up in, so the top of any crate is what has actually been picked rather than what has the most stars.