skills.rest
linenoize
linenoizeCOMMUNITY
Explore all skills by this author →
GitHub

context-pack

Creates structured handoff packets with task constraints and acceptance criteria for delegated AI work.

1|Updated May 18, 2026
Search Tags:#subagents#acceptance criteria#task delegation#context handoff#briefing#durable context
One-click install
npx skills add https://github.com/linenoize/topia --skill context-pack-linenoize
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: context-pack
Source: https://github.com/linenoize/topia/tree/main/skills/context-pack
Command: npx skills add https://github.com/linenoize/topia --skill context-pack-linenoize

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

Context-pack prevents delegated work from losing decisions, constraints, and prior attempts when an AI agent hands a task to another agent or resumes work later. It turns scattered conversation history into a compact briefing that is easier for the next agent to use than rereading the full thread.

Core Features & Use Cases

  • Structured handoff packets: Captures task goals, decisions, constraints, and progress in a predictable format.
  • Durable context summaries: Focuses on type names, contracts, and behavior instead of brittle file paths or line numbers.
  • Delegation support: Helps build, team, rescue, and other higher-level workflows spawn subagents without repeating work.
  • Safety checks: Enforces mandatory sections like out-of-scope boundaries and testable acceptance criteria.
  • Use case: A coding assistant splits a large refactor into subtasks and uses this Skill to brief each subagent so they stay aligned.

Quick Start

Use context-pack to turn the current task, constraints, decisions, and failed attempts into a compact handoff brief for the next agent.

Frequently Asked Questions about context-pack

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

FAQPage Schema
How do I preserve context when handing off tasks to AI subagents?▼

Context handoff for subagents requires structured packets that capture type-and-contract summaries, mandatory out-of-scope boundaries, and behavior-based acceptance criteria. This durable context approach prevents agents from relying on brittle file paths or line numbers.

How do I split complex tasks for subagents without losing decisions and constraints?▼

Split complex tasks by turning scattered conversation history into a compact handoff brief that enforces mandatory sections like out-of-scope boundaries and testable acceptance criteria. This ensures spawned subagents stay aligned with parent agent decisions.

What should a context handoff packet include for delegated AI work?▼

A context handoff packet should include durable type-and-contract summaries, mandatory out-of-scope boundaries, and behavior-based acceptance criteria. These elements ensure the next agent receives essential task context within a compact prompt budget.

When do I need durable context summaries for AI task delegation?▼

You need durable context summaries when an AI agent splits large refactors into subtasks, spawns subagents, or passes work across sessions. This prevents the next agent from rereading the full thread to understand prior attempts.

Why does my AI subagent repeat work or ignore constraints after a session handoff?▼

Subagents repeat work after a session handoff because scattered conversation history loses decisions and constraints. Structuring handoff packets with explicit out-of-scope boundaries and behavior-based acceptance criteria prevents this context loss.

Related Skills

Explore More Software Engineering Skills→
TheycallmehollaTheycallmeholla

handoff

Compacts active conversation context into a structured handoff document.

Community
Intermediate
special-place-administratorspecial-place-administrator

session-handoff

Summarize active work into a compact HANDOFF block for the next session.

Community
Intermediate
shuff57shuff57

hermes-agent-orchestration

Orchestrate AI agents in sequential and parallel workflows.

Community
Advanced
HuuBarHuuBar

continue-claude-work

Extract session metadata and summaries from Claude Code artifacts into a resume briefing.

Community
Advanced