rysweet/amplihack-rs/context-management
Claude Code and agent skill by rysweet/amplihack-rs
- community
Checking that the SKILL.md file is still in the repository.
Proactive context window management via token monitoring, intelligent extraction, and selective rehydration. Features context health checks, priority-based retention, and recovery guidance. Use when approaching token limits or needing to preserve essential context. Complements /transcripts, /reflect, session checkpoints, and PreCompact hooks.
Install
npx vskill@latest install rysweet/amplihack-rs/context-managementSKILL.md
Context Management Skill
Purpose
Preserve the information needed to continue work across long sessions, compaction, crashes, and handoffs. This skill uses the active agent runtime plus a bundled native shell helper for status and snapshot management.
When to Use
- The conversation is getting long or noisy.
- A task spans multiple phases, repos, or agents.
- You need a compact handoff after a crash or compaction.
- You need to rehydrate only requirements, decisions, files touched, validation state, and blockers.
Workflow
- Assess context health: identify active goal, constraints, changed files, open todos, validation status, and unresolved questions.
- Extract essentials: keep user requirements, decisions, file paths, commands run, failures, fixes, and next actions.
- Drop noise: summarize verbose logs and omit repeated command output unless it contains an unresolved error.
- Checkpoint: write persistent notes through the active session checkpoint/transcript mechanism when available, or use the bundled helper.
- Rehydrate: restore only the minimum context needed for the next action, then verify against repository state before continuing.
Output Shape
Use this structure for handoffs:
Native Helper
Use the bundled helper for repeatable context status and snapshots:
Snapshots are written to ${AMPLIHACK_CONTEXT_DIR:-$HOME/.amplihack/context-management}.
Native Validation
Use runtime commands and repository checks alongside the helper:
If those commands are unavailable, produce the same handoff structure manually from the current conversation and repository state.
Security
- Security scanPassed, 90 of 100Pattern checks on the skill's files.
- Model reviewPassed
Quality evaluation
- Task evaluationNot runNo evaluation yet. Run one in Skill Studio.
Source
- SKILL.md in the repositoryCheckingamplifier-bundle/skills/context-management/SKILL.md
Popularity
- Repository stars
- 11
- Repository forks
- 7
- 7-day trend score
- 0
Stars and forks are for the whole GitHub repository, not this skill alone. Not refreshed yet.
Works with
Any agent vskill installs to: 56 targets in all. These 39 have a page on this site.
- Claude Code
- Cursor
- GitHub Copilot
- Windsurf
- Codex
- Gemini CLI
- Cline
- Amp
- Roo Code
- Goose
- Continue
- Kilo Code
- Junie
- Trae
- OpenHands
- Qwen Code
- AI Agent
- Tabnine Chat
- Sourcegraph Cody
- Amazon Q
- Firebase Genkit
- Sweep
- Devin
- Phind
- Mentat
- Claude Desktop
- Aider
- GPT Pilot
- Plandex
- bolt.new
- v0
- Replit Agent
- Anse
- TypingMind
- BetterChatGPT
- LibreChat
- Warp AI
- Pieces
- Fig AI
Embed badge
[](https://verified-skill.com/skills/rysweet/amplihack-rs/context-management)