skills.rest
the-agency-ai
the-agency-aiOFFICIAL
Explore all skills by this author →
WebsiteGitHub

compact-resume

Verify session state, relaunch dead monitors, and list unread dispatches after compaction.

39|6|Updated Jan 5, 2026
Search Tags:#resume#handoff#session#monitor#compact#session-pickup
One-click install
npx skills add https://github.com/the-agency-ai/the-agency --skill compact-resume
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: compact-resume
Source: https://github.com/the-agency-ai/the-agency/tree/main/.claude/skills/compact-resume
Command: npx skills add https://github.com/the-agency-ai/the-agency --skill compact-resume

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

This Skill enables automatic continuation of a conversation session after it has been compacted, ensuring that context, state, and monitoring are verified and restored for smooth resumption.

Core Features & Use Cases

  • Handoff Reading & Verification: Reads the existing handoff file, checks the session state, and verifies that the working tree is clean for a safe resume.
  • Monitor Re-Initialization: Detects dead monitors and automatically relaunches them to restore full system health.
  • Dispatch Drift Analysis: Lists and displays any message dispatches received during the compact process to inform the user before resuming.
  • Use Case: Ideal for ongoing AI-assisted development sessions that require pausing and resuming without losing context or missing critical updates.

Quick Start

Run ./agency/tools/session-pickup --from compact and review the output; then continue your task seamlessly from the verified state.

Frequently Asked Questions about compact-resume

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

FAQPage Schema
How do I resume a development session after a conversation compact without losing context?▼

Resuming a session after compacting involves verifying the session state and working tree, re-launching dead monitors, and checking unread dispatches to ensure no critical updates are missed during the transition.

What is conversation compaction and when do I need to verify session state?▼

Conversation compaction compresses dialogue history to free up context. You need session state verification afterward to ensure the working tree is clean, monitors are active, and no dispatch messages were lost during compression.

How do I check for missed message dispatches after a context window compaction?▼

To check for missed message dispatches after a context window compaction, this Skill performs a dispatch drift analysis that lists and displays any messages received during the compact process before you continue working.

Why are my background monitors dead after resuming an AI-assisted coding session?▼

Background monitors often die during conversation compression. This Skill detects dead monitors and automatically re-initializes them to restore full system health before resuming your AI-assisted coding session.

Can I automatically restore session monitors and clean working tree state after pausing?▼

Yes, you can automatically restore session monitors and verify a clean working tree state by reading the existing handoff file, which checks session health and relaunches necessary monitors for a safe resume.

How do I use session-pickup to continue work from a compacted state?▼

To use session-pickup, run the pickup script with the compact flag to read the handoff file, verify the working tree, and review output for unread dispatches before seamlessly continuing your task.

Related Skills

Explore More Software Engineering Skills→
htlin222htlin222
77

update-dev-docs

Sync development documentation with current TodoWrite state and implementation progress.

Community
Intermediate
KorytoKoryto
3

init

Initialize a dotagent session by claiming session state with a valid session_id.

Community
Intermediate
the-agency-aithe-agency-ai
39

compact-prepare

Create structured session handoff files for multi-agent workflow compaction.

Official
Intermediate
GrimblazGrimblaz
3

subagent-env-handshake

Verify live repository state via a six-field Markdown-HTML handshake.

Community
Intermediate