meerkat-architecture

Document Meerkat internal architecture, crate ownership, and DSL runtime responsibilities.

19|5|Updated Jan 22, 2026
One-click install
npx skills add https://github.com/lukacf/meerkat --skill meerkat-architecture
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: meerkat-architecture
Source: https://github.com/lukacf/meerkat/tree/main/.claude/skills/meerkat-architecture
Command: npx skills add https://github.com/lukacf/meerkat --skill meerkat-architecture

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Provides an internal architecture guide for the Meerkat agent platform, helping developers understand crate ownership, trait contracts, the agent construction pipeline, session lifecycle, runtime control plane, mob orchestration internals, and comms wiring. It is oriented toward AI agents and developers working on meerkat internals, not end users.

Core Features & Use Cases

  • Serves as a navigational map to architecture docs and references for cross-crate coordination.
  • Clarifies crate ownership and DSL vs runtime responsibilities to prevent shadow semantics.
  • Aids new contributors and maintainers in aligning changes with architectural dogma and canonical schemas.

Quick Start

Read this guide to quickly locate internal architecture references and cross-crate details you need.

Frequently Asked Questions about meerkat-architecture

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

FAQPage Schema
How do I navigate Meerkat's internal architecture across multiple crates?

To navigate Meerkat's internal architecture, use this guide to locate cross-crate coordination references, crate ownership boundaries, and trait contracts. It serves as a navigational map for developers reasoning about runtime responsibilities and DSL bindings.

What is the difference between DSL and runtime responsibilities in Meerkat?

Meerkat separates DSL and runtime responsibilities to prevent shadow semantics. DSL bindings define canonical machine schemas, while the runtime control plane manages session lifecycle, agent construction, and mob orchestration execution.

How does mob orchestration work within the Meerkat runtime?

Mob orchestration in the Meerkat runtime coordinates multi-agent execution through the control plane. It relies on comms wiring and cross-crate access patterns to ensure consistent behavior across orchestrated agents and sessions.

Do I need to understand trait contracts before extending Meerkat crates?

Yes, understanding trait contracts is essential before extending Meerkat crates. The architecture guide documents canonical schemas and cross-crate access patterns that ensure new contributions align with architectural dogma and prevent inconsistent behavior.

Why does Meerkat enforce canonical machine schemas across crates?

Meerkat enforces canonical machine schemas across crates to prevent shadow semantics and ensure architecture alignment. Documenting these schemas guarantees that DSL bindings and runtime control plane interactions produce consistent behavior throughout the agent platform.