Agent Plugins Marketplace
← All plugins

context-management

v0.3.0

Skills for keeping Claude context lean, accurate, and correctly registered. Start with optimize, the roll-up entry point: it audits a scope (CLAUDE.md/AGENTS.md, skills, memory), fixes redundant/contradictory/dead instructions directly, and surfaces candidates for the specialist skills it hands off to: extract-to-skill (pull misplaced-weight content into its own skill), skill-registry-audit (find duplicate or broken skill registrations), memory-to-context (promote team-useful project memories into tracked project docs), and memory-hygiene (dedupe and prune stale memory in place). Each specialist also triggers standalone.

Claude Code5 Skills

By Mark Labrecque0 GitHub starsUpdated last week

Directory evidence

Runtimes
Claude Code
Parsed components
5 skill or MCP entries
Source updated
Oct 8, 2026
Manifest status
Canonical path parsed

The directory validates manifest shape and source location. It does not execute the plugin or provide a security endorsement. Review the indexing methodology →

Install context-management for Claude Code

Installs for the current user
claude plugin marketplace add IchenDEV/agent-plugin-mkt
claude plugin marketplace update agent-plugin-marketplace
claude plugin install context-management-4@agent-plugin-marketplace

Paste and run these commands in a terminal with Claude Code. They add and refresh the PluginsMP catalog, then install this plugin.

The installer fetches third-party code from the source repository shown on this page. This directory validates manifest structure and source location, but does not perform a security audit; review the manifest, components, and source before installing.

Get the source manually
git clone https://github.com/markdlabrecque/personal-ai-plugins

Clone the source repository, then follow its setup instructions to add the plugin to a compatible client. The plugin root is context-management/.

Plugin files

context-management/
├── .claude-plugin/plugin.json
├── skills/extract-to-skill/SKILL.md
├── skills/memory-hygiene/SKILL.md
├── skills/memory-to-context/SKILL.md
├── skills/optimize/SKILL.md
└── skills/skill-registry-audit/SKILL.md

Included Skills5

extract-to-skillskills/extract-to-skill/SKILL.md

Pull a large or rarely-needed block out of an always-loaded context file (CLAUDE.md/AGENTS.md) into its own skill, loaded on demand instead of paying its token cost every session. Use whenever the user says "extract this to a skill", "make this a skill", "pull this out", "this is too big for CLAUDE.md", or confirms an optimize skill's misplaced-weight finding. Also trigger when a section is clearly a self-contained, occasionally-needed procedure (a runbook, a workflow with an embedded script, reference material for one narrow scenario) sitting in an always-loaded file.

memory-hygieneskills/memory-hygiene/SKILL.md

Clean up a project's (or every project's) Claude memory in place — find memory files that duplicate or contradict each other, and project-status memories that have gone stale (past due dates, closed tickets, superseded decisions) and should be archived or updated. Use whenever the user asks to clean up memory, audit memory, prune stale memories, or mentions memory files that seem out of date or repetitive. Distinct from memory-to-context, which promotes team-useful memories into project docs — this skill is about the memory store's own internal health, not what should move to the repo.

memory-to-contextskills/memory-to-context/SKILL.md

Promote Claude memory files with durable, team-wide value (codebase facts, architecture, conventions, decisions) into the project's tracked docs — AGENTS.md, README, docs/ — then remove them from memory. Use when the user wants memories promoted to docs or memory content visible to teammates. Never promote personal workflow preferences; those stay in memory.

optimizeskills/optimize/SKILL.md

Entry point for auditing and cleaning up Claude context — CLAUDE.md/AGENTS.md, their @-imports, skills, and memory in a scope. Fixes redundant, contradictory, or dead instructions directly and routes bigger findings to the other context-management skills. Use on "optimize context", "audit / clean up CLAUDE.md or instructions", "find contradictions in my rules". Takes scope "project" or "global"; ask if unstated.

skill-registry-auditskills/skill-registry-audit/SKILL.md

Audit the skill registry itself — skills registered in more than one place, settings entries enabling missing plugins, installed-but-never-enabled plugins, and SKILL.md vs marketplace.json drift. Use on "audit skills", "find duplicate skills", "clean up plugins", or when a registered skill isn't triggering. (optimize audits instruction text; this audits the registry.)

Plugin manifests1

context-management/.claude-plugin/plugin.json
{
  "name": "context-management",
  "version": "0.3.0",
  "description": "Skills for keeping Claude context lean, accurate, and correctly registered. Start with optimize, the roll-up entry point: it audits a scope (CLAUDE.md/AGENTS.md, skills, memory), fixes redundant/contradictory/dead instructions directly, and surfaces candidates for the specialist skills it hands off to: extract-to-skill (pull misplaced-weight content into its own skill), skill-registry-audit (find duplicate or broken skill registrations), memory-to-context (promote team-useful project memories into tracked project docs), and memory-hygiene (dedupe and prune stale memory in place). Each specialist also triggers standalone.",
  "author": {
    "name": "Mark Labrecque",
    "email": "[email protected]"
  }
}

For maintainers

If you maintain this plugin, link to this source-backed listing from your README so users can review its manifest and indexed components.

[context-management on Agent Plugins Marketplace](https://pluginsmp.com/plugins/context-management-4)