Codenamev Claude Memory
codenamev-claude_memory is a Claude Code plugin for the Build phase that provides journey-wide, SQLite-backed long-term memory via hooks and MCP tools.
by codenamev · github.com/codenamev/claude_memory
Give Claude Code durable, self-managed memory across sessions using hooks, MCP tools, and a local SQLite store with provenance tracking.
Add it to Claude Code
Install the plugin in Claude Code. One command, paste-ready.
/plugin install codenamev-claude_memory@codenamev/claude_memoryBuilt to be called by your agent
Skillselion is itself an MCP server. Your agent can pull this entry and a paste-ready install config straight from the API - no copy-paste.
Retrieve this entry with skillselion.get_details("plugin:codenamev/claude_memory") and the paste-ready config with skillselion.get_install_config("plugin:codenamev/claude_memory").
What it does
codenamev-claude_memory is a Claude Code plugin that adds long-term, self-managed memory through hooks, MCP tools, and SQLite so solo builders do not restart context from zero every session. It targets builders who treat the coding agent as a long-running collaborator on one product: decisions, constraints, and facts should persist with provenance instead of living only in yesterday’s chat export. Hooks automate when memory is written or refreshed; MCP exposes structured operations the agent can call; SQLite keeps data local and under your control. Expect to invest setup time in maintenance and extraction behavior—the listing stresses truth and tracking, which implies you still curate what gets remembered. Install during Build when you wire agent infrastructure, then rely on it across Idea research recall, shipping fixes, and Operate iteration without re-explaining your stack each time.
Highlights
- Long-term, self-managed memory layer for Claude Code beyond a single chat thread
- Hooks and MCP tools coordinate capture, extraction, and maintenance workflows
- SQLite-backed local store with provenance and truth-oriented tracking keywords
- Designed for ongoing memory upkeep rather than manual copy-paste session notes
- Single-plugin marketplace entry focused on memory—not a general skill library
Why builders use it
Claude Code forgets project decisions and constraints between sessions, so you waste time re-teaching context the agent should already know.
After setup, important facts and provenance can persist in a self-managed local memory store the agent can read and maintain across sessions.
At a glance
- Type - Plugin in AI Agents.
- Adoption - 0 installs, 16 stars, 0 votes.
FAQ
Who is codenamev-claude_memory for?
Indie developers using Claude Code on multi-week projects who need durable, locally controlled memory with extraction and provenance—not ephemeral thread context.
When should I use codenamev-claude_memory?
Use it as soon as agent setup begins and keep it on across the journey whenever session continuity, tracked facts, or memory maintenance matter.
How do I add codenamev-claude_memory to my agent?
Install the codenamev/claude_memory plugin marketplace entry in Claude Code, configure hooks and the MCP memory server, and ensure SQLite storage paths are writable on your machine.
Comments
Share how you use codenamev-claude_memory, gotchas, or tips for other indie builders.
No comments yet - be the first to share how you use it.