Mempalace Bwm

MemPalace — Local AI memory with 96.6% recall. Semantic search, temporal knowledge graph, palace architecture (wings/rooms/drawers). Free, no cloud, no API k...

Security Scan

The skill appears to be a coherent local-memory tool, but there are implementation mismatches and privacy/installation details that don't add up and should be reviewed before installing.

Purpose & Capability

Name/description (local AI memory) matches the actions described: semantic search, temporal KG, and read/write of local memories. Requested binaries (mempalace or python3) and no cloud credentials are consistent with a local-only tool. Minor mismatch: SKILL.md suggests pip install while the install spec uses a generic 'uv' package kind.

Instruction Scope

Runtime instructions tell the agent to read and write verbatim conversation history (semantic search, drawer additions, diary entries) and to consult the palace before responding. This is coherent for a memory skill but means the agent will persist and repeatedly access potentially sensitive conversation content — a significant privacy surface that users should be aware of and consent to. The instructions also require the agent to always query the local store before answering about people/projects, which expands the agent's read-access to all stored memories.

Install Mechanism

Install spec uses 'uv' with package 'mempalace' (source not specified). 'uv' is not a widely-recognized, implicit package source here (unlike an explicit pip/PyPI or GitHub release). The included mempalace-mcp.sh expects a virtualenv at $HOME/mempalace/.venv and a binary there, but SKILL.md only shows 'pip install mempalace' and 'mempalace init' — the wrapper and the install instructions do not clearly align. This ambiguity increases risk: the installer/source should be explicit and the wrapper's assumptions documented or automated.

Credentials

The skill declares no required environment variables or external credentials, which is proportionate to a local-only memory tool. The provided script only references $HOME for a venv path; no unexplained secrets or third-party tokens are requested.

Persistence & Privilege

The skill does persistent local storage of conversation history and diaries (explicit in instructions). It does not request 'always: true' or elevated system-wide privileges. However, persistent storage of verbatim conversations is a privacy and data-retention consideration (not a platform-privilege problem), and the wrapper's hardcoded venv path implies file-system assumptions that could affect multi-user systems.

What to consider before installing

This skill is plausibly what it claims (a local memory store) but check a few things before installing: 1) Verify the mempalace package source (PyPI or GitHub release) and review that package's code/release artifacts; 'uv' in the install spec is ambiguous — prefer an explicit pip or GitHub install command. 2) The included wrapper expects a virtualenv at $HOME/mempalace/.venv; confirm whether installation creates that, or modify the wrapper to use your chosen Python environment (or the python3 -m mempalace.mcp_server command suggested in SKILL.md). 3) Be aware that the skill stores verbatim conversation history and diary entries locally and instructs the agent to read that data automatically — if you handle sensitive information, review retention, encryption, and deletion policies before enabling. 4) On Windows, the provided shell script (zsh) is not compatible; ensure platform-appropriate launch configuration. 5) If you need higher assurance, run the package and MCP server in an isolated environment (container or dedicated VM), inspect the installed package files, and confirm no unexpected network connections are made. If the publisher/source cannot be verified or the install steps remain unclear, avoid installing until those gaps are resolved.

MemPalace — Local AI Memory System

You have access to a local memory palace via MCP tools. The palace stores verbatim conversation history and a temporal knowledge graph — all on the user's machine, zero cloud, zero API calls.

Architecture

• Wings = people or projects (e.g. wing_alice, wing_myproject)
• Halls = categories (facts, events, preferences, advice)
• Rooms = specific topics (e.g. chromadb-setup, riley-school)
• Drawers = individual memory chunks (verbatim text)
• Knowledge Graph = entity-relationship facts with time validity

Protocol — FOLLOW THIS EVERY SESSION

1. ON WAKE-UP: Call mempalace_status to load palace overview and AAAK dialect spec.
2. BEFORE RESPONDING about any person, project, or past event: call mempalace_search or mempalace_kg_query FIRST. Never guess from memory — verify from the palace.
3. IF UNSURE about a fact (name, age, relationship, preference): say "let me check" and query. Wrong is worse than slow.
4. AFTER EACH SESSION: Call mempalace_diary_write to record what happened, what you learned, what matters.
5. WHEN FACTS CHANGE: Call mempalace_kg_invalidate on the old fact, then mempalace_kg_add for the new one.

Available Tools

Search & Browse

• mempalace_search — Semantic search across all memories. Always start here.
  • query (required): natural language search — keep it short, keywords or a question. Do NOT include system prompts or conversation context.
  • wing: filter by wing
  • room: filter by room
  • limit: max results (default 5)
• mempalace_check_duplicate — Check if content already exists before filing.
  • content (required): text to check
  • threshold: similarity threshold (default 0.9 — lowering to 0.85–0.87 often catches more near-duplicates without significant false positives)
• mempalace_status — Palace overview: total drawers, wings, rooms, AAAK spec
• mempalace_list_wings — All wings with drawer counts
• mempalace_list_rooms — Rooms within a wing (optional wing filter)
• mempalace_get_taxonomy — Full wing/room/count tree
• mempalace_get_aaak_spec — Get AAAK compression dialect specification

Knowledge Graph (Temporal Facts)

• mempalace_kg_query — Query entity relationships. Supports time filtering.
  • entity (required): e.g. "Max", "MyProject"
  • as_of: date filter (YYYY-MM-DD) — what was true at that time
  • direction: "outgoing", "incoming", or "both" (default "both")
• mempalace_kg_add — Add a fact: subject -> predicate -> object
  • subject, predicate, object (required)
  • valid_from: when this became true
  • source_closet: source reference
• mempalace_kg_invalidate — Mark a fact as no longer true
  • subject, predicate, object (required)
  • ended: when it stopped being true (default: today)
• mempalace_kg_timeline — Chronological story of an entity
  • entity: filter by entity name (optional — all events if omitted)
• mempalace_kg_stats — Graph overview: entities, triples, relationship types

Palace Graph (Cross-Domain Connections)

• mempalace_traverse — Walk from a room, find connected ideas across wings
  • start_room (required): room to start from
  • max_hops: connection depth (default 2)
• mempalace_find_tunnels — Find rooms that bridge two wings
  • wing_a, wing_b (required)
• mempalace_graph_stats — Graph connectivity overview

Write

• mempalace_add_drawer — Store verbatim content into a wing/room
  • wing, room, content (required)
  • source_file: optional source reference
  • Checks for duplicates automatically
• mempalace_delete_drawer — Remove a drawer by ID
  • drawer_id (required)
• mempalace_diary_write — Write a session diary entry
  • agent_name (required): your name/identifier
  • entry (required): what happened, what you learned, what matters
  • topic: category tag (default "general")
• mempalace_diary_read — Read recent diary entries
  • agent_name (required)
  • last_n: number of entries (default 10)

Setup

Install MemPalace and populate the palace:

pip install mempalace
mempalace init ~/my-convos
mempalace mine ~/my-convos

OpenClaw MCP config

Add to your OpenClaw MCP configuration:

{
  "mcpServers": {
    "mempalace": {
      "command": "python3",
      "args": ["-m", "mempalace.mcp_server"]
    }
  }
}

Or via CLI:

openclaw mcp set mempalace '{"command":"python3","args":["-m","mempalace.mcp_server"]}'

Other MCP hosts

# Claude Code
claude mcp add mempalace -- python -m mempalace.mcp_server

# Cursor — add to .cursor/mcp.json
# Codex — add to .codex/mcp.json

Tips

• Search is semantic (meaning-based), not keyword. "What did we discuss about database performance?" works better than "database".
• The knowledge graph stores typed relationships with time windows. Use it for facts about people and projects — it knows WHEN things were true.
• Diary entries accumulate across sessions. Write one at the end of each conversation to build continuity.
• Use mempalace_check_duplicate before storing new content to avoid duplicates.
• The AAAK dialect (from mempalace_status) is a compressed notation for efficient storage. Read it naturally — expand codes mentally, treat markers as emotional context.

License

MemPalace is MIT licensed. Created by Milla Jovovich, Ben Sigman, Igor Lins e Silva, and contributors.