Production Agent Builder

Structured 8-step framework for building production AI agents. Use when designing a new AI agent, planning agent architecture, building an automated workflow...

Security Scan

This is a coherent, instruction-only design framework for building production AI agents; it does not request credentials, install software, or perform unexpected actions.

Purpose & Capability

The name and description match the SKILL.md and references/guide.md content: both provide an 8-step framework for designing production agents. The guidance about inputs, tools, memory, safeguards, interfaces, and testing is appropriate for an 'agent builder' skill.

Instruction Scope

The instructions stay within the stated purpose — they are a planning/design framework. They recommend connecting to external tools (email, Slack, DBs, vector stores) and logging tool calls, which is expected for a design guide but implies implementers will later need to attach credentials and tooling. The SKILL.md does not itself instruct reading system files, environment variables, or sending data to hidden endpoints.

Install Mechanism

No install spec and no code files are present (instruction-only). That minimizes risk because nothing will be written to disk or fetched automatically during install.

Credentials

The skill declares no required environment variables or credentials, which is proportional. However, its recommendations (attach data tools, action tools, vector stores, etc.) mean that any real implementation of the design will require external API keys/credentials; users should apply least-privilege and provide only the necessary credentials at implementation time.

Persistence & Privilege

always is false and the skill is user-invocable. It does not request persistent presence, system configuration changes, or access to other skills' configs. The default ability for an agent to invoke the skill autonomously is normal and not, by itself, a concern here.

Assessment

This skill is a guidance document (no code, no installs) and appears internally consistent with its stated purpose. Before using it in production: (1) verify the source/author (no homepage is provided), (2) when you implement the design, provide only the minimal credentials required for each connector (apply least privilege), (3) enforce the suggested safeguards — gate high-risk actions behind human approval and enable auditing/logging, and (4) review any platform-specific integrations (email, Slack, DBs, vector stores) for where data will be stored or transmitted to avoid unintended data exposure.

AI Agent Builder

Structured framework for building AI agents that work in production. Based on the Storm & Storm methodology.

When to Use

• Designing a new AI agent from scratch
• Planning architecture for an automated workflow
• Reviewing or improving an existing agent's design
• Teaching someone how to build agents

The 8-Step Process

Follow these steps in order. Each step has a clear goal and concrete deliverables.

Step 1: Choose a Task

Pick ONE painful, repeating workflow. Not "AI in general."

• Must be repeatable (weekly+), follow steps, have clear I/O
• Define success: "Given X, the agent should output Y so that Z happens."

Step 2: Map the Steps

Break the task into 4–7 steps: INPUT → ACTIONS → DECISION → OUTPUT

• Classify each step: ⚖️ pure rules | 📖 heavy reading/writing | 🎯 judgement calls
• Choose infrastructure (no-code vs dev-friendly)
• You need: strong model + tool calling + basic logs

Step 3: Specify Inputs, Outputs & Tools

Treat the agent like an API, not a chatbot.

• Define required input fields (text, file, URL, ID)
• Define structured outputs (JSON/template the system can trust)
• Attach tools: data (search/DB/CRM), action (email/Slack/tasks), orchestration (schedulers/webhooks/queues)

Step 4: Write the System Prompt

Create a clear role with: role definition, boundaries, style, 1–2 example conversations.

• Use ReAct pattern: observe → think → act → reflect

Step 5: Add Memory

Three layers: conversation state, task memory, knowledge memory (vector store/file search).

• Key question: "What does this agent need to remember for the next step to be smarter?"

Step 6: Add Safeguards

Gate high-risk actions (email, data changes, money) behind human approval.

• Rules: never invent IDs, ask when ambiguous
• Log every tool call and decision for audit

Step 7: Build the Interface

Match to where users work: chat, Slack command, button in app, or web form.

Step 8: Test

For each real example: watch the trace, score correctness + efficiency + time saved.

• Tighten prompts/tools/rules where it fails. Iterate.

Detailed Reference

For expanded details on each step, including selection criteria, classification examples, tool categories, memory layer patterns, and a pre-launch checklist:

→ Read references/guide.md

Output Format

When using this framework to design an agent, produce a design document covering:

# Agent Design: [Name]

## Task & Success Criteria
[Step 1 output]

## Step Map
[Step 2 output — numbered steps with classifications]

## I/O Specification
[Step 3 output — inputs, outputs, tools]

## System Prompt
[Step 4 output — the actual prompt]

## Memory Architecture
[Step 5 output — which layers, what's stored]

## Safeguards
[Step 6 output — gated actions, rules, logging]

## Interface
[Step 7 output — chosen interface and why]

## Test Plan
[Step 8 output — example inputs, expected outputs, scoring criteria]