How do I install the Context skill?

Run npx skillmds add majiayu000/context-6 in your terminal (requires Node.js), paste this page's agent-chat prompt into Claude, Cursor, or any MCP-connected agent, or download the SKILL.md file and copy it into your agent's skills directory.

Is Context safe to use?

This skill has not completed SkillMD's automated safety review yet. Capability flags: reads secrets. SkillMD never runs a skill's scripts for you; review the SKILL.md before installing.

Which AI agents work with Context?

This skill is tagged as working with Claude Code, Claude.ai, OpenAI Codex. SKILL.md is an open format, so most agents that read a skills directory can load it too.

Is Context free to use?

Yes. Installing skills from SkillMD is free, and the skill stays under its author's original license.

Who published Context?

majiayu000 (@majiayu000) published this skill. Their other Agent Skills are listed on their SkillMD profile.

Context by majiayu000 — Agent Skill

Name: Context
Rating: 4.2 (5 reviews)
Author: majiayu000

Context

Context window management. Use when: - User asks about "context window", "context usage", "context limit" - User mentions "running out of context", "context full" - User wants to "compact", "clear context", "optimize context" - Context is getting full and needs management

majiayu000 f166730 2 files · 2.8 KB Updated Sep 9, 2026 567 repo stars

File contents

Context Window Management

Monitor and manage Claude Code context window usage.

Check Context Usage

Run /context in Claude Code to see current usage.

Context Thresholds

Thresholds scale with your context window size:

Default Context (~200K tokens)

Level	Usage	Action
Notice	70% (~140K)	Consider handoff at next stopping point
Warning	80% (~160K)	Create handoff, prepare to compact
Critical	90% (~180K)	Create handoff NOW, then compact

Managing Context

Create Handoff Before Compacting

/create-handoff

This saves:

Current task state
Key decisions made
Files modified
Next steps

Compact Context

/compact

Reduces context by summarizing earlier conversation.

Resume After Compact

/resume-handoff

Restores context from handoff file.

Tips for Reducing Context

Use context: fork - Skills that fork don't bloat main context
Delegate exploration - Let agents handle research
Clean summaries - Agents return summaries, not raw output
Avoid large file reads - Use TLDR for token-efficient analysis

When Context is Full

Don't panic - Your work isn't lost
Create handoff - Save current state
Compact - Let Claude summarize
Resume - Pick up where you left off

Automatic Protection

The setup includes:

Warning at 70%, 80%, 90% thresholds
Auto-handoff before compaction (PreCompact hook)
Session start recalls relevant learnings

Output

Report:

Current usage: X% of context
Recommendation: Action to take
Next steps: What to do

---name: context-6description: Context window management. Use when: - User asks about "context window", "context usage", "context limit" - User mentions "running out of context", "context full" - User wants to "compact", "clear context", "optimize context" - Context is getting full and needs management--- # Context Window Management Monitor and manage Claude Code context window usage. ## Check Context Usage Run `/context` in Claude Code to see current usage. ## Context Thresholds Thresholds scale with your context window size: ### Default Context (~200K tokens) | Level | Usage | Action | |-------|-------|--------| | Notice | 70% (~140K) | Consider handoff at next stopping point | | Warning | 80% (~160K) | Create handoff, prepare to compact | | Critical | 90% (~180K) | Create handoff NOW, then compact | ## Managing Context ### Create Handoff Before Compacting ``` /create-handoff ``` This saves: - Current task state - Key decisions made - Files modified - Next steps ### Compact Context ``` /compact ``` Reduces context by summarizing earlier conversation. ### Resume After Compact ``` /resume-handoff ``` Restores context from handoff file. ## Tips for Reducing Context 1. **Use `context: fork`** - Skills that fork don't bloat main context 2. **Delegate exploration** - Let agents handle research 3. **Clean summaries** - Agents return summaries, not raw output 4. **Avoid large file reads** - Use TLDR for token-efficient analysis ## When Context is Full 1. **Don't panic** - Your work isn't lost 2. **Create handoff** - Save current state 3. **Compact** - Let Claude summarize 4. **Resume** - Pick up where you left off ## Automatic Protection The setup includes: - Warning at 70%, 80%, 90% thresholds - Auto-handoff before compaction (PreCompact hook) - Session start recalls relevant learnings ## Output Report: - **Current usage**: X% of context - **Recommendation**: Action to take - **Next steps**: What to do

Frequently asked questions