Use when active context is large, stale, contradictory, repeatedly reread, or no longer represents current source and task state.
Scanned 9/6/2026
Install to Claude Code
npx -y skills add ihabkhaled/AI-Psychiatry --skill context-refresh --agent claude-codeInstalls into .claude/skills of the current project.
Are you the author of Context Refresh?
Add the live security badge to your README — it updates automatically with every re-scan.
[](https://www.skillsdirectory.com/skills/ihabkhaled-context-refresh)More formats (shields.io, HTML) on the badges page.
---
name: context-refresh
description: Use when active context is large, stale, contradictory, repeatedly reread, or no longer represents current source and task state.
---
# Context Refresh
## Core principle
Refresh preserves facts, decisions, blockers, and state while dropping speculation and reloading only changed authoritative sources.
## Procedure
1. Summarize goal, completed, remaining, decisions, evidence, blocker, and next action.
2. Discard abandoned hypotheses and repetitive output.
3. Check source hashes, versions, or modification evidence.
4. Reload only stale or changed sources.
5. Resolve conflicts in favor of current source.
6. Update compact task context and resume.
## Required output
Compressed state, refreshed sources, retired stale facts, and next action.
## Limits
Two identical reloads without source change trigger attention reset.
## Common mistakes
Do not reread the entire repository or trust stale summaries over source.
## Stop condition
Stop when current context is small, fresh, and sufficient for the next action.
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.