agentforge-spec

Generate a formal Agent Spec from five Phase 0 questions.

Updated Apr 8, 2026

Search Tags:#workflow #architecture #agent #feasibility #spec #phase-0

Ready to run

npx skills add https://github.com/Kingxiao/agentforge --skill agentforge-spec

💬 Prompt for ChatGPT / Claude Chat▼

Please help me install this Agent Skill.
Skill: agentforge-spec
Source: https://github.com/Kingxiao/agentforge/tree/main/agentforge-spec
Command: npx skills add https://github.com/Kingxiao/agentforge --skill agentforge-spec

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Phase 0 defines whether your use case should be solved by an Agent or a deterministic workflow, and outputs the formal Agent Spec to guide architecture.

Core Features & Use Cases

Determines agent necessity using the Suitability Decision Tree and feasibility checks.
Guides Phase 1 handoff with a structured Agent Spec document that includes user inputs and automated inferences (Agent Type, Interaction Pattern, Architecture Paradigm, etc.).
Reduces risk by locking critical requirements (security baseline, memory strategy, tool interfaces) before any code is written.

Quick Start

Answer the five Phase 0 questions to generate the Agent Spec.

Frequently Asked Questions about agentforge-spec

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

FAQPage Schema

How do I decide if my use case needs an AI agent or a deterministic workflow?▼

To decide between an AI agent and a deterministic workflow, you apply a Suitability Decision Tree and feasibility checks. This gating process evaluates whether your task requires autonomous decision-making or fixed, rule-based execution paths to mitigate architectural risk.

What is an Agent Spec document and when do I need it for architecture?▼

An Agent Spec is a formal document defining user intent, technical constraints, and acceptance criteria for an AI agent. You need it during Phase 0 to lock critical requirements like security baselines and memory strategy before writing any code.

How do I define security baselines and memory strategy before building an AI agent?▼

You define security baselines and memory strategy by answering five Phase 0 questions about your use case. The resulting structured output captures these technical constraints and tool interfaces to lock requirements before development begins.

What inputs are required to generate a formal Agent Spec for a Phase 1 handoff?▼

Generating a formal Agent Spec requires user-provided answers to five specific Phase 0 questions. These inputs enable the system to infer details like Agent Type, Interaction Pattern, and Architecture Paradigm for the downstream Phase 1 handoff.

Can I use this process to validate agent feasibility and mitigate project risks?▼

Yes, you can use this process to validate agent feasibility and mitigate risks. The Phase 0 gating decisions identify unsuitable use cases early and enforce the definition of tool interfaces and acceptance criteria prior to code implementation.