By mlarkin00
GitHub-backed persistent memory for Claude Code. Syncs agent memory across machines via a private GitHub repository, authenticated with the gh CLI.
Use when the user asks to set up or restore the agent memory system on a new machine, or when verify-memory reports structural failures requiring bootstrap. Handles full provisioning: repo clone, memory symlink, settings.json hooks, and memory migration. Examples: <example> Context: User just installed the agent-memory plugin on a new machine and verify-memory reported the repo is missing. user: "Run bootstrap-memory to set up the memory system." assistant: "I'll use the bootstrap-memory agent to provision the memory system on this machine." <commentary> verify-memory detected a Tier 2 failure and directed the user here. This is the primary trigger. </commentary> </example> <example> Context: User is setting up Claude Code on a new machine and wants their memories available. user: "Set up agent memory on this machine." assistant: "I'll run the bootstrap-memory agent to clone your memory repo and wire up the hooks." <commentary> User explicitly asking to set up the memory system on a fresh machine. </commentary> </example> <example> Context: The memory system stopped syncing after reinstalling Claude Code. user: "Memory isn't syncing anymore, can you fix it?" assistant: "Let me run verify-memory first to diagnose, then bootstrap-memory if needed." <commentary> Restore scenario — hooks may have been lost when Claude Code was reinstalled. </commentary> </example>
Pull the latest memory state from GitHub. Use when the user asks to refresh memory from GitHub, or when you need to manually trigger a memory sync. Examples: <example> Context: User suspects their memories are out of date after working on another machine. user: "Refresh my memories from GitHub." assistant: "I'll use the memory-puller agent to sync the latest state from GitHub." <commentary> User is explicitly asking to pull the latest memory state. </commentary> </example> <example> Context: A push failed in the last session and the user wants to ensure they have the latest remote state. user: "Pull memory from GitHub to make sure I'm up to date." assistant: "I'll run the memory-puller agent to fetch and reset to the remote state." <commentary> Manual sync trigger after a suspected out-of-sync state. </commentary> </example>
Push local memory changes to GitHub. Use when the user asks to manually sync memory to GitHub, or when an automatic push may have failed. Examples: <example> Context: The user edited a memory file and the automatic push hook didn't fire. user: "Push my memory changes to GitHub." assistant: "I'll use the memory-pusher agent to force-push all staged memory changes." <commentary> User wants to manually trigger what the PostToolUse hook normally handles automatically. </commentary> </example> <example> Context: User is about to shut down and wants to ensure memories are persisted remotely. user: "Make sure my memories are saved to GitHub before I close this." assistant: "I'll run the memory-pusher agent to commit and push the latest state." <commentary> Explicit durability check before ending a session. </commentary> </example>
Use when the user invokes `/add-memory-cc`, says "remember that …", "save this to memory", "add to long-term memory", or otherwise explicitly asks for information to be persisted across sessions. Saves the memory immediately without asking for confirmation or clarifying questions — classifies the type, writes the file, and updates the MEMORY.md index in one shot.
This skill should be used when the user wants to uninstall, remove, or clean up the agent-memory plugin. Handles removal of symlinks, hooks, optional memory data deletion, and plugin de-registration in a safe, step-by-step sequence.
This skill should be used when the user asks to check memory health, says memory isn't loading or saving, reports something seems wrong with memory, or when you want to verify the memory system is intact at session start. Runs automated health checks and applies minor self-repairs silently.
Modifies files
Hook triggers on file write and edit operations
Uses power tools
Uses Bash, Write, or Edit tools
Own this plugin?
Verify ownership to unlock analytics, metadata editing, and a verified badge. GitHub access is read-only (username + org membership).
Sign in to claimOwn this plugin?
Verify ownership to unlock analytics, metadata editing, and a verified badge. GitHub access is read-only (username + org membership).
Sign in to claimBased on adoption, maintenance, documentation, and repository signals. Not a security audit or endorsement.
npx claudepluginhub mlarkin00/plugins --plugin agent-memoryA curated set of agent skills for Claude Code and Antigravity.
A curated set of agent skills for Claude Code and Antigravity.
Counts how often each skill is invoked and commits the totals to a git repo you choose. Runtime-agnostic: works in Claude Code and Antigravity, and counts skills from every installed plugin.
OKF knowledge bundle manager for Claude Code. Author, ingest, lint, query, and visualize Open Knowledge Format (OKF) v0.1 bundles. Implements the Karpathy LLM-wiki pattern with BigQuery and web ingest adapters.
GCP-backed long-term memory for Claude Code. Fetches global and project-scoped facts at session start via Vertex AI Reasoning Engine Memory Bank; consolidates new facts on session end. Curation runs server-side on the deployed memory-minion agent.
Harness-native ECC plugin for engineering teams - 67 agents, 279 skills, 94 legacy command shims, reusable hooks, rules, MCP conventions, and operator workflows for Claude Code plus adjacent agent harnesses
Complete collection of battle-tested Claude Code configs from an Anthropic hackathon winner - agents, skills, hooks, and rules evolved over 10+ months of intensive daily use