Archetypal Ai
Give long-running coding agents durable recall, remember, and checkpoint memory across sessions.
Overview
io.github.archetypal-ai/archetypal-ai is a MCP server for the Build phase that provides persistent recall, remember, and checkpoint memory for AI coding agents.
What is this MCP server?
- npm package archetypal-ai with stdio MCP transport
- Tools themed around recall, remember, and checkpoint for soul preservation across sessions
- GitHub source at archetypal-ai/archetypal-ai
- Registry version 1.0.5 on the official MCP server schema
- npm package version 1.0.5
- stdio transport via identifier archetypal-ai
What problem does it solve?
Every new agent session forgets your architecture choices, tone, and half-finished tasks.
Who is it for?
Indie builders running multi-day Claude Code or Cursor projects who need durable agent memory.
Skip if: One-off scripts or teams forbidden from local persistence without a reviewed data policy.
What do I get? / Deliverables
After install, agents can remember, recall, and checkpoint context so work continues across sessions with less re-onboarding.
- Working recall, remember, and checkpoint MCP tools in the agent
- Cross-session continuity for decisions and task state
Recommended MCP Servers
Journey fit
Persistent agent memory is installed while building agent stacks, even though it pays off across Ship review and Operate iteration. recall, remember, and checkpoint are core agent-tooling primitives, not a single SaaS integration.
How it compares
Agent memory MCP server, not a brainstorming or planning methodology skill.
Common Questions / FAQ
Who is io.github.archetypal-ai/archetypal-ai for?
Solo builders and small teams building long-lived AI agents who need recall and checkpoints beyond a single chat window.
When should I use io.github.archetypal-ai/archetypal-ai?
Early in Build when you adopt agent-tooling, and ongoing through Ship and Operate whenever session continuity matters.
How do I add io.github.archetypal-ai/archetypal-ai to my agent?
Install the npm package archetypal-ai and register stdio MCP transport in Claude Code, Cursor, or Codex per your client’s MCP config format.