spec

Generate standardized feature spec templates for AI agent session continuity.

Updated Aug 23, 2026
One-click install
npx skills add https://github.com/kpeez/agentspec --skill spec-kpeez
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: spec
Source: https://github.com/kpeez/agentspec/tree/main/skills/spec
Command: npx skills add https://github.com/kpeez/agentspec --skill spec-kpeez

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Teams building AI agents often struggle to keep context across sessions. This skill provides a reproducible framework to create and maintain per-feature specs that preserve goals, decisions, and context for ongoing work.

Core Features & Use Cases

  • Scaffold per-feature specs with standard templates (AGENTS.md, design.md, ledger.md, decisions.md, future-work.md).
  • Enable quick handoffs by consolidating context and decisions into a single, versioned folder under specs/<feature>.
  • Support future extensions by providing a consistent template structure for feature work.

Quick Start

Run /spec new <name> to scaffold the feature-spec template and initialize per-feature context.

Frequently Asked Questions about spec

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

FAQPage Schema
How do I maintain context continuity for AI agents across sessions?

You can maintain context continuity by creating structured feature specs that consolidate goals, decisions, and history into versioned folders under specs/<feature>. This preserves ongoing work context for new sessions.

What templates do I need to document feature specs for agent handoffs?

Feature spec documentation requires standard templates including AGENTS.md, design.md, ledger.md, decisions.md, and future-work.md. These files scaffold per-feature context to enable quick agent handoffs.

How do I scaffold a new feature spec folder?

Run /spec new <name> to initialize a feature-spec template folder. This command populates standardized scaffolding under specs/<feature> to support frontmatter-driven discovery and future expansion.

When should I create per-feature specs for AI agent workflows?

Create per-feature specs when starting a new feature or updating existing specifications. This ensures context continuity across agent sessions by preserving design decisions and goals in a reproducible framework.

Can I customize the standardized spec templates for future expansion?

Yes, the scaffolding supports optional sections designed for future expansion. You can populate the standard templates with frontmatter-driven discovery while extending the structure for specific per-feature work needs.

Why use structured feature specs instead of ad-hoc documentation for agent context?

Structured feature specs provide a reproducible framework that consolidates context and decisions into a single versioned folder. Ad-hoc documentation lacks this standardization, making agent handoffs and context recovery unreliable.