agent-scaffolder

Scaffold AgentStack agents from trigger to deployment with vault and KB seeding.

3|3|Updated Mar 5, 2026
One-click install
npx skills add https://github.com/webdevcom01-cell/agent-studio --skill agent-scaffolder
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: agent-scaffolder
Source: https://github.com/webdevcom01-cell/agent-studio/tree/main/skills/agent-scaffolder
Command: npx skills add https://github.com/webdevcom01-cell/agent-studio --skill agent-scaffolder

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This Skill automates the end-to-end scaffolding of a new AgentStack agent, transforming a user trigger into a production-ready agent with a defined DESIGN_SPEC, AgentStack entry, standard flow, vault files, KB seed, and a smoke test.

Core Features & Use Cases

  • Fully extracts agent name from triggers and collects a minimal spec (pipeline role, model, web data requirement)
  • Generates processor and extractor prompts with real output keys, then writes DESIGN_SPEC to the vault
  • Creates the AgentStack agent and builds a standard 3-5 node flow (kb_search → processor → extractor, plus optional web_search and call_agent)
  • Initializes vault documents (evo-log, instincts, and design spec), seeds a KB with 3 documents, and patches the kb_search node
  • Performs a smoke test via as_chat_with_agent and reports scaffold outcomes

Quick Start

Trigger the scaffold with a clear agent description and let it generate DESIGN_SPEC, create the agent, build the flow, initialize vault, seed the KB, and run a smoke test.

Frequently Asked Questions about agent-scaffolder

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I scaffold a production-grade AI agent in AgentStack?

AgentStack agent scaffolding handles trigger-to-deployment by generating DESIGN_SPEC, creating the agent, building a 3-5 node flow, initializing vault documents, and running a smoke test.

What does automated agent scaffolding include for end-to-end flow generation?

Automated agent scaffolding includes generating processor and extractor prompts, writing DESIGN_SPEC to the vault, creating a standard flow with KB search and optional web search, and seeding the knowledge base.

How do I generate a DESIGN_SPEC and initialize a vault for a new AI agent?

Generate a DESIGN_SPEC and initialize a vault by triggering the scaffold with an agent description, which automatically extracts the name, collects minimal specs, and writes evo-log, instincts, and design spec files.

Does automated agent scaffolding enforce model selection and safety guardrails?

Automated agent scaffolding enforces explicit model selection, A2A wiring rules, and safety guardrails to prevent misconfigurations during the AgentStack agent creation and flow construction process.

What is the best way to verify AI agent readiness after scaffolding a flow?

The best way to verify AI agent readiness is through the built-in smoke test, which uses as_chat_with_agent to validate the scaffolded flow, vault, and KB seed outcomes before deployment.

Can I add web search and call agent nodes when scaffolding an AgentStack flow?

Yes, you can add optional web_search and call_agent nodes alongside the standard kb_search, processor, and extractor nodes when scaffolding an AgentStack flow with web data requirements.