MemPalace AI memory — persistent memory across sessions. Search past decisions, store architecture choices, track temporal network facts via knowledge graph, navigate cross-domain connections, maintain specialist agent diaries. Use when recalling past decisions, storing important context, tracking network changes over time, or maintaining operational journals.
Instalar com Codex ou Claude Copie este prompt, cole no Codex, Claude ou outro assistente e deixe que ele revise a página da skill e instale para você.
Um comando direto ignora o prompt de revisão. Verifique a origem antes de executá-lo.
Instruções da origem · Visualização somente leitura
name
mempalace
description
MemPalace AI memory — persistent memory across sessions. Search past decisions, store architecture choices, track temporal network facts via knowledge graph, navigate cross-domain connections, maintain specialist agent diaries. Use when recalling past decisions, storing important context, tracking network changes over time, or maintaining operational journals.
Relationship to OpenClaw's built-in memory: OpenClaw writes raw daily logs to memory/YYYY-MM-DD.md. MemPalace adds a structured, searchable layer on top — semantic search across all sessions, a temporal knowledge graph for network facts, and per-agent diaries. Use daily logs for "what happened today" and MemPalace for "what did we decide and why."
How to Call the Tools
All tools use mcp-call with the mempalace MCP server:
Returns total drawers, wing/room counts, palace path, memory protocol instructions, and AAAK dialect spec. Call this at session start to load palace context.
python3 $MCP_CALL"python3 -u $MEMPALACE_MCP_SCRIPT" mempalace_add_drawer '{"wing":"wing_netclaw","room":"architecture-decisions","content":"Campus OSPF design: Area 0 backbone on core switches, Area 10 Building A, Area 20 Building B. Stub areas on access layer."}'
File verbatim content into the palace. Checks for duplicates first.
Parameters:
wing (required): Wing name (project/person)
room (required): Room name (topic/aspect)
content (required): Verbatim content to store — exact words, never summarized
Add a fact to the knowledge graph. Subject → predicate → object with optional time window. E.g. ('Max', 'started_school', 'Year 7', valid_from='2026-09-01').
Query the knowledge graph for an entity's relationships. Returns typed facts with temporal validity. Filter by date with as_of to see what was true at a point in time.
Parameters:entity (required), as_of (optional, YYYY-MM-DD), direction (optional: outgoing/incoming/both).
Walk the palace graph from a room. Shows connected ideas across wings — the tunnels. Like following a thread through the palace: start at 'chromadb-setup' in wing_code, discover it connects to wing_myproject (planning).
Write to your personal agent diary in AAAK format. Your observations, thoughts, what you worked on, what matters. Each agent has their own diary with full history. Write in AAAK for compression — e.g. 'SESSION:2026-04-04|built.palace.graph+diary.tools|ALC.req:agent.diaries.in.aaak|★★★'. Use entity codes from the AAAK spec.
python3 $MCP_CALL"python3 -u $MEMPALACE_MCP_SCRIPT" mempalace_add_drawer '{"wing":"wing_netclaw","room":"routing-decisions","content":"Migrated campus core from OSPF to eBGP. Reason: multi-vendor support (Arista + Cisco). AS 65001 for core, AS 65010-65020 for distribution."}'
