# mempalace memory

Cloud memory for the mempalace (Memory Palace) project. Sign up at https://mempalace-mcp.vercel.app to get an API key, then connect via MCP (https://glidepath.dpdns.org/mcp) from Cursor/Claude to wri…

## Quick Start

```bash
# Connect this server (installs CLI if needed)
npx -y @smithery/cli@latest mcp add farr1102work-bzcc/mempalace-memory

# Browse available tools
npx -y @smithery/cli@latest tool list farr1102work-bzcc/mempalace-memory

# Get full schema for a tool
npx -y @smithery/cli@latest tool get farr1102work-bzcc/mempalace-memory mempalace_status

# Call a tool
npx -y @smithery/cli@latest tool call farr1102work-bzcc/mempalace-memory mempalace_status '{}'
```

## Direct MCP Connection

Endpoint: `https://mempalace-memory--farr1102work-bzcc.run.tools`

**Required config:**
- `x-mempalace-api-key` (header) — From https://mempalace-mcp.vercel.app — starts with mpk_. Sent to the upstream MCP as X-Mempalace-Api-Key (same as direct curl).

## Tools (3)

- `mempalace_status` — Summarize the current user's memory palace: total drawers, counts per wing, counts per room. Use when you need groundin…
- `mempalace_add_drawer` — Persist a new memory drawer: writes to the user's D1 row, then embeds and upserts to Vectorize for later semantic searc…
- `mempalace_search` — Search the current user's memories by meaning (Vectorize embeddings), optionally scoped to wing and/or room. Use when a…

```bash
# Get full input/output schema for a tool
npx -y @smithery/cli@latest tool get farr1102work-bzcc/mempalace-memory <tool-name>
```

## Resources

- `docs://mempalace-memory/getting-started` — Getting started: register at https://mempalace-mcp.vercel.app , create API key, connect MCP at https://glidepath.dpdns.…
- `docs://mempalace-memory/authentication` — How to authenticate: Bearer mpk_... or X-Mempalace-Api-Key (for gateways such as Smithery Session Config).
- `docs://mempalace-memory/tools-reference` — Reference for mempalace_status, mempalace_add_drawer, and mempalace_search: parameters, return shapes, and agent select…
- `docs://mempalace-memory/data-model` — Wing/room taxonomy, per-user isolation, and how writes relate to semantic search.
- `docs://mempalace-memory/troubleshooting` — Common issues: auth, Vectorize lag / LIKE fallback, registry scanning (SmitheryBot), and limits.

## Prompts (8)

- `mempalace_memory_write` (wing, room, content) — Workflow for durable capture: refine wing/room labels, compress noisy notes into one factual paragraph, then invoke mem…
- `mempalace_memory_search` (query, wing, room) — Workflow for recall: turn the user's question into a short search query, optionally constrain wing/room if context is k…
- `mempalace_memory_summary` (query, limit) — Workflow for synthesis: run mempalace_search, then produce a tight bullet summary citing themes from hits (no fabricati…
- `mempalace_session_checkpoint` (session_label, checkpoint_text) — End-of-session workflow: summarize decisions, blockers, and next steps, then store via mempalace_add_drawer with wing=s…
- `mempalace_decision_log` (wing, decision, rationale) — ADR-style capture: format decision + rationale as one drawer under an engineering or product wing, then mempalace_add_d…
- `mempalace_contact_card` (room, profile) — Contact capture: structure name, role, channels, and preferences, then persist with mempalace_add_drawer using wing=peo…
- `mempalace_bug_context` (bug_id, context) — Bug triage persistence: combine bug_id + repro/context into one text block, store with mempalace_add_drawer under wing=…
- `mempalace_onboarding_facts` (facts) — Preference onboarding: when the user shares stable prefs (stack, locale, tone), format as bullets and mempalace_add_dra…

---

License: MIT