wrap

Persist session state and target mutations before clearing conversation context.

Updated Feb 22, 2026
One-click install
npx skills add https://github.com/marcelocantos/skills --skill wrap-marcelocantos
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: wrap
Source: https://github.com/marcelocantos/skills/tree/main/skills/wrap
Command: npx skills add https://github.com/marcelocantos/skills --skill wrap-marcelocantos

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Pre-clear housekeeping to reliably capture what changed during a session, update target state, and persist the minimal context needed for the next convergence review so a fresh session can resume without loss.

Core Features & Use Cases

  • Safety-net drafting: immediately persist a short wrap-draft to auto-memory to protect against context exhaustion.
  • Target management: propose and apply target mutations (status changes, creations, reframings, weight changes, retirements) via bullseye MCP tools so docs/targets.md is rendered by the single source of truth.
  • Context persistence: save debugging insights or topic files to auto-memory and write a concise ## Last session section to MEMORY.md for immediate load in subsequent conversations.
  • Cleanup: remove the transient wrap-draft when work is complete to avoid stale interrupted-wrap state.
  • Use case: run at the end of a coding or design session to snapshot work completed, update project targets, and prepare for the next /cv without manual bookkeeping.

Quick Start

Run the wrap skill to save a short draft, apply bullseye target updates, write a Last session entry into MEMORY.md, and clean up the transient draft.

Frequently Asked Questions about wrap

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

FAQPage Schema
How do I save conversation context and session state before clearing context?

Save session state before clearing by persisting a concise summary into MEMORY.md and applying target updates, which captures in-flight work and learnings so a fresh session can resume without loss.

What is the best way to update project targets before ending an AI coding session?

Update project targets before ending a session by applying target mutations like status changes, creations, and retirements via bullseye MCP tools, ensuring docs/targets.md reflects a single source of truth.

How does safety-net drafting protect against context exhaustion during session management?

Safety-net drafting protects against context exhaustion by immediately persisting a short wrap-draft to auto-memory, creating an atomic snapshot of in-flight work before the final conversation context clearing occurs.

Can I use this session wrap workflow for end-of-session convergence reviews?

Yes, this workflow is designed for end-of-session convergence reviews, recording targets, in-flight work, and learnings to prepare the minimal context needed for the next review.

Why does a transient wrap draft need cleanup after persisting memory?

Transient wrap drafts require cleanup after persisting memory to remove stale interrupted-wrap state, ensuring subsequent sessions do not accidentally load outdated or incomplete session snapshots.