skills.rest
mikeparcewski
mikeparcewskiCOMMUNITY
Explore all skills by this author →
WebsiteGitHub

smaht

Gathers requested context from wicked-brain, wicked-garden, and domain state on demand.

8|2|Updated Feb 9, 2026
Search Tags:#context#subagents#briefing#pull-model#wicked-brain#domain-state
One-click install
npx skills add https://github.com/mikeparcewski/wicked-garden --skill smaht
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: smaht
Source: https://github.com/mikeparcewski/wicked-garden/tree/main/skills/smaht
Command: npx skills add https://github.com/mikeparcewski/wicked-garden --skill smaht

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Gathers relevant on-demand context from wicked-brain + wicked-garden:search and domain state when a subagent or command asks for it, providing a concise briefing before a task and reducing prompt churn.

Core Features & Use Cases

  • Pulls background context when a subagent or command asks for it
  • Keeps context up-to-date with domain state without pushing it on every prompt
  • Aliases: context-assembly, briefing, smart-context

Quick Start

Request a briefing with "assemble context" or "give me a briefing" to pull the latest background for the current task.

Frequently Asked Questions about smaht

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

FAQPage Schema
How do I get context briefing for subagents without pushing it on every prompt?▼

A pull-model context briefing fetches relevant background from domain state and memory stores only when a subagent requests it. This on-demand assembly prevents prompt churn by avoiding pushing static context with every single prompt invocation.

What is the difference between push and pull context assembly for AI agents?▼

Pull context assembly retrieves background information on-demand when a subagent requests a briefing, whereas push models inject context into every prompt automatically. The pull approach reduces prompt churn and keeps context aligned with current domain state.

How do I assemble background context from domain state before starting a task?▼

Request a briefing using commands like "assemble context" or "give me a briefing" to pull the latest relevant background from your domain state and memory stores. This provides a concise context briefing immediately before task execution.

Can I use a pull-model to fetch domain state for subagents dynamically?▼

Yes, a pull-model supports fetching domain state dynamically when subagents or commands request it. This allows subagents to operate with up-to-date context assembled on-demand rather than relying on pre-pushed static information.

Why does pushing context with every prompt increase prompt churn?▼

Pushing context with every prompt increases churn because it repeatedly injects static background regardless of relevance. A pull-model solves this by gathering context only on-demand during task briefing, minimizing unnecessary token usage and keeping context current.

Related Skills

Explore More Software Engineering Skills→
FacundoSu1986FacundoSu1986

metacognitive-context-orchestrator

Orchestrate metacognitive reasoning workflows with local vector stores and state machines.

Community
Advanced
SterlingChinSterlingChin
1.0k

start

Load MARVIN context and goals to deliver a daily briefing.

Community
Basic
BeltDoorBeltDoor

begin-session

Load client, project, and initiative context into concise operational briefings.

Community
Advanced
shuu5shuu5

spawn

Spawn a new tmux window and execute a prompt without prior context.

Community
Advanced