Use when choosing which repository instructions, architecture, domain knowledge, files, tests, or deep guidance to load for a task without reading everything.
Scanned 9/6/2026
Install to Claude Code
npx -y skills add ihabkhaled/AI-Psychiatry --skill context-router --agent claude-codeInstalls into .claude/skills of the current project.
Are you the author of Context Router?
Add the live security badge to your README — it updates automatically with every re-scan.
[](https://www.skillsdirectory.com/skills/ihabkhaled-context-router)More formats (shields.io, HTML) on the badges page.
---
name: context-router
description: Use when choosing which repository instructions, architecture, domain knowledge, files, tests, or deep guidance to load for a task without reading everything.
---
# Context Router
## Core principle
Load minimal sufficient context in layers: boot, task index, domain, exact rules and files, then deep architecture only when required.
## Procedure
1. Classify the task type and required decision.
2. Load the tiny bootstrap and task index.
3. Select one relevant domain and its test commands.
4. Load exact rules and sources needed for the active work item.
5. Escalate to deep architecture only for a proven dependency.
6. Record the route and avoid duplicate sources.
## Required output
Task type, context layers loaded, canonical sources, and freshness.
## Limits
Ordinary tasks should rarely use deep layer L4.
## Common mistakes
Do not read every AI or architecture document before locating the relevant source.
## Stop condition
Stop when context is sufficient for reliable action and additional reading has no completion value.
Is this your skill, or is something wrong with this listing? Request removal or report an issue. Author removals are honored within 72 hours.
No comments yet. Be the first to comment!
Ultra-compressed communication mode. Cuts token usage ~75% by speaking like caveman while keeping full technical accuracy. Supports intensity levels: lite, full (default), ultra, wenyan-lite, wenyan-full, wenyan-ultra. Use when user says "caveman mode", "talk like caveman", "use caveman", "less tokens", "be brief", or invokes /caveman. Also auto-triggers when token efficiency is requested.