skills.rest
anthony-chaudhary
anthony-chaudharyCOMMUNITY
Explore all skills by this author →
WebsiteGitHub

memory-compact

Compact Claude Code memory stores into hot and cold tiers with Python integrity checks.

30|12|Updated Jun 21, 2026
Search Tags:#optimization#maintenance#memory#archiving#data-integrity#claude-code
One-click install
npx skills add https://github.com/anthony-chaudhary/fak --skill memory-compact-anthony-chaudhary
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: memory-compact
Source: https://github.com/anthony-chaudhary/fak/tree/main/.claude/skills/memory-compact
Command: npx skills add https://github.com/anthony-chaudhary/fak --skill memory-compact-anthony-chaudhary

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This skill resolves the silent data loss occurring when the Claude Code auto-memory store exceeds the harness load cap, causing older, critical memory entries to become invisible to the agent.

Core Features & Use Cases

  • Structural Tiering: Automatically splits memory into a hot tier (active context) and a cold tier (archived history) to ensure essential data always loads.
  • Integrity Witnessing: Uses a dedicated script to verify that the memory index remains within line and byte limits while maintaining a perfect bijection between index references and disk files.
  • Use Case: When the agent reports that memory is full or truncating, use this skill to re-tier the index and verify that no data has been orphaned or lost.

Quick Start

Run the memory-compact skill to audit the current memory store and identify entries that should be moved to the archive tier.

Frequently Asked Questions about memory-compact

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

FAQPage Schema
Why does Claude Code memory become full and stop loading older entries?▼

Memory tiering splits the Claude Code auto-memory store into a hot tier for active context and a cold tier for archived history. This structural separation ensures essential data always loads within harness limits while older entries are preserved.

Do I need Python 3 to run memory compaction and integrity checks for Claude Code?▼

You can verify memory index integrity by running a deterministic witness script that requires Python 3. This script validates that the memory index maintains a perfect bijection with disk files and remains within line and byte limits, ensuring no data is lost.

How do I verify memory index integrity and check for orphaned files in Claude Code?▼

You can verify memory index integrity by running a deterministic witness script that requires Python 3. This script validates that the memory index maintains a perfect bijection with disk files and remains within line and byte limits, ensuring no data is lost.

What is the best way to archive old context and manage memory limits in Claude Code?▼

The best way to manage memory limits is to use structural tiering to automatically migrate older memory entries from the hot tier to a cold archive tier. This maintains the harness load cap and preserves historical data without silent truncation.

When should I audit and re-tier my Claude Code memory directory?▼

You should audit and re-tier your memory directory when the agent reports that memory is full or truncating, or when critical older memory entries become invisible to the agent. Re-tiering moves entries to the archive tier to verify no data has been orphaned.

Related Skills

Explore More Software Engineering Skills→
gcudergcuder
3

harness-generator

Generate agent harnesses with linters, style enforcement, and plan gates.

Community
Advanced
Fr-e-dFr-e-d
157

memory-index-sync

Detect and resolve drift between index.md and files on disk.

Community
Intermediate
ChegginCheggin
1

tiered-memory

Manage hot, warm, and cold memory tiers on disk with TTLs.

Community
Advanced
KYRIE66nbKYRIE66nb

note

Save timestamped context notes to .omx/notepad.md across sessions.

Community
Intermediate