pyrs-spec

Create and maintain persistent pyramid specification files with YAML frontmatter and Markdown.

2|Updated Apr 6, 2026

Search Tags:#context management #specification #AI coding #code architecture #pyramids

Ready to run

npx skills add https://github.com/zachbutton/pyrs --skill pyrs-spec

💬 Prompt for ChatGPT / Claude Chat▼

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

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires pyrs-foundation, and includes scripts (resource) and references (resource) components.

What problem does it solve?

The lack of persistent context in AI coding agent sessions leads to lost intent, architectural drift, and fragile changes. PYRS prevents these issues by producing persistent markdown files ("pyramids") that capture system behavior.

Core Features & Use Cases

Persistent Context: Captures system behavior and intent across sessions.
Pyramid Creation: Generates new pyramids for concepts or revises existing ones.
Spec Commands: Uses ::spec to introduce new concepts or update existing ones.
Audit Loop: Ensures pyramids stay aligned with the actual system.

Quick Start

Invoke the pyrs-foundation skill, then use ::spec @.event-bus — a centralized pub/sub messaging layer to create a pyramid for a new concept.

Frequently Asked Questions about pyrs-spec

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

FAQPage Schema

How do I maintain persistent context across AI coding agent sessions?▼

Persistent context in AI coding agents is maintained by generating markdown pyramid specifications that capture system behavior. These files prevent architectural drift and lost intent by preserving behavioral state across sessions.

What is the best way to prevent architectural drift when using AI coding tools?▼

Preventing architectural drift in AI coding tools is best handled by maintaining persistent markdown specifications that align with system behavior. An audit loop ensures these specifications remain accurate as the underlying system changes.

How do I create a new specification for a code architecture concept?▼

To create a new specification for a code architecture concept, invoke the foundation context management skill and use the spec command with a concept description. This generates a structured markdown pyramid with YAML frontmatter.

Do I need a context management dependency to generate markdown specifications?▼

Yes, generating markdown specifications requires a context management foundation dependency. The pyramid specification process relies on this prerequisite to handle context creation and revision operations for AI coding agents.

How does an audit loop keep code architecture specifications aligned with behavior?▼

An audit loop keeps code architecture specifications aligned by continuously comparing the persistent markdown pyramids against actual system behavior. This process identifies behavioral drift and revises the specifications to match the current state.

Can I update an existing specification without losing the original system intent?▼

Updating an existing specification without losing original system intent is possible through targeted pyramid revision. The process modifies the YAML frontmatter and markdown structure while preserving the core architectural behavior captured in the file.