nap

Compresses, prunes, and archives agent histories and state in .squad/.

2|Updated Apr 21, 2026
One-click install
npx skills add https://github.com/elbruno/md-to-slides --skill nap-elbruno
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: nap
Source: https://github.com/elbruno/md-to-slides/tree/main/.copilot/skills/nap
Command: npx skills add https://github.com/elbruno/md-to-slides --skill nap-elbruno

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Long-running multi-agent sessions can exceed the context window because agent histories, logs, and stale state keep accumulating.

Core Features & Use Cases

  • Compress agent histories to reduce token usage while preserving the essentials needed for continued work.
  • Prune old logs so obsolete events don’t waste context budget.
  • Archive stale decisions to retain important outcomes without keeping everything in hot context.
  • Clean orphaned inbox files to remove leftover artifacts after tasks complete.

Use case: after a sprint where many agents spawned and context grew rapidly, run a cleanup to bring .squad/ state back under budget so the next round of planning and execution can proceed.

Quick Start

Run squad nap --deep to aggressively compress, prune, and archive stale .squad/ state.

Frequently Asked Questions about nap

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

FAQPage Schema
How do I reduce context window usage when multi-agent histories grow too large?

Context window usage is reduced by compressing multi-agent histories, pruning old logs, and archiving stale decisions. This removes obsolete events and orphaned inbox artifacts from hot context while preserving essential outcomes needed for continued work.

When should I prune and archive accumulated state in a long-running agent session?

Pruning and archiving accumulated state is needed when agent histories, logs, and stale decisions keep accumulating and exceed practical context window limits. It is especially necessary before heavy fan-out operations and after long multi-agent runs.

Can I safely dry-run context cleanup on a large .squad directory before applying changes?

Safe dry-run behavior is supported for context cleanup on a large .squad directory. This allows you to preview scoped cleanup of histories, logs, stale decisions, and orphaned inbox artifacts before any actual compression or archiving occurs.

What's the best way to aggressively compress and archive stale multi-agent state?

Aggressive compression and archiving of stale multi-agent state is achieved by running a deep mode cleanup. This deeply compresses agent histories, prunes obsolete logs, and archives stale decisions to bring .squad state back under budget.

Does context compression preserve essential multi-agent session outcomes?

Context compression preserves essential outcomes needed for continued work while removing obsolete events. Important stale decisions are archived rather than deleted, retaining necessary history without keeping everything in hot context budget.

Why does my multi-agent history.md file keep growing beyond practical size limits?

Multi-agent history.md files grow beyond practical size because agent histories, logs, and stale state continuously accumulate during long-running sessions. Running scoped cleanup compresses and prunes these accumulated artifacts to reclaim context budget.