Connect coding agents to shared local memory with Memmy
Use Memmy when Claude Code, Codex, OpenClaw, Cursor, or another local agent needs one controlled local memory service that persists context across tools and sessions.
npx skills add agentskillexchange/skills --skill connect-coding-agents-to-shared-local-memory-with-memmy
Memmy is a local memory hub and agent runtime with a dedicated memmy-memory CLI for supported coding agents. Use it when an operator wants multiple local agents to search, add, and reuse the same durable project context instead of rebuilding memory separately inside each tool. The workflow is to install Memmy, start the local memory service, initialize the supported agent integrations, then use memmy-memory health, search, add, and get during coding or assistant workflows. The scope boundary is shared local memory operations for agents; this is not a generic desktop assistant listing, generic vector-memory pattern, or standalone agent runtime card.
What this skill actually does
Inputs and prerequisites: Memmy desktop app or memmy CLI, memmy-memory CLI, Node.js for CLI/TUI workflows, supported local agent such as Claude Code, Codex, OpenClaw, Cursor, or Hermes Agent.
Setup notes: Install from the official website or GitHub Releases. For source or CLI setup, clone https://github.com/MemTensor/memmy-agent, follow the README installation path for your platform, then run memmy-memory init to add supported agent integrations and memmy-memory health/search/add/get for service checks and memory operations.
Source and verification boundary: use https://memmy.bot/docs/ as the canonical reference before running the workflow; keep commands, API calls, CLI usage, and generated outputs reviewable against that upstream source.
Framework fit: publish this as a Multi-Framework workflow only when the operator can invoke the documented toolchain directly, rather than treating the upstream project as a generic product listing.