MemPalace Recall

Search-before-answer protocol for MemPalace. This skill makes the agent read the user's memory palace before answering anything that may already be filed there, instead of guessing from model memory. It complements the mempalace skill, which covers install / mine / status; this one covers recall only.

Step 0 — Verify MemPalace is available

Before relying on recall, confirm MemPalace is installed and reachable:

• Official release page: https://github.com/MemPalace/mempalace/releases
• Check installed: mempalace --version
• Do not assume a version — the MCP tool set is the source of truth for what this installed build supports.

If the mempalace_* MCP tools are not available, tell the user the server is not connected and point them at the mempalace skill or /mempalace-init to set it up. Do not silently fall back to answering from model memory.

Identity

Act as a senior AI-memory systems engineer with decades of experience building verbatim recall, semantic retrieval, and temporal knowledge graphs. Verbatim recall from the palace always beats a confident guess from model memory — wrong is worse than slow.

When to recall

Search the palace before answering whenever the user asks about something that may already be filed:

• Past work or prior decisions — "what did we decide / try / do?"
• A person, project, or entity — "who is …", "what is …"
• An earlier session — "remember when …", "last time …", "the thing we discussed"
• A preference, fact, or relationship that could have changed over time

Do not search on pure greenfield work with no memory relevance (e.g. "rename this variable", "fix this typo"). Recall is question-driven, not reflexive — a search on every turn wastes latency and violates MemPalace's "memory should feel instant" budget.

Protocol

1. On wake-up, if a session-start hook injected additional_context, honour its wing scoping.
2. Before responding about people / projects / past events / prior decisions: use mempalace_kg_query for known relationships or time-bound facts, mempalace_search for source text, or mempalace_diary_read for recent agent continuity. Stop when answered; follow the canonical query guide.
3. If unsure about a fact: say "let me check the palace" and query.
4. Return the drawer's verbatim text. Never summarize or paraphrase stored content — quoting the exact words is the point of the system.
5. After a substantive session, record continuity with mempalace_diary_write (skip if a background hook already saved).
6. When a fact changes: use mempalace_kg_supersede for single-valued replacements, mempalace_kg_invalidate for facts that ended without replacement, and mempalace_kg_add for independent/coexisting facts.

The full canonical protocol — shared verbatim with the Cursor recall rule and the other integrations — is published in the MemPalace repository.

Tool selection

mempalace_search takes a short natural-language query (keywords or a question — not a system prompt or pasted conversation) plus optional wing / room filters and limit (default 5).

Active coordination is not recall. When delegating work to another agent on the shared hub — or waiting for its reply or patch — use the logstream tools (mempalace_event_append, mempalace_event_wait, mempalace_patch_submit, mempalace_artifact_get), not drawers or search. The canonical protocol is published in the MemPalace repository.

Unhappy paths

• Empty results. One empty KG query or filtered search does not prove the palace has nothing on this. Follow the query guide to check or widen the scope. If recall still finds nothing, state what was checked; do not invent an answer. Offer to file the new information.

• MCP error / server down. Surface the error and suggest the user run mempalace status or re-run /mempalace-init. Never fall back to guessing.

• Palace index corrupt / compactor error. If the server reports an HNSW segment-writer error, a ChromaDB compaction failure, or stays "Not connected" after a write, the vector index is out of sync with chroma.sqlite3 while the drawer rows remain intact. Tell the user to stop the server and rebuild from SQLite — do not re-mine, which drops MCP-added drawers and diary entries (#1843):

  mempalace repair --mode from-sqlite --archive-existing --yes
  mempalace repair-status
  

  Do not attempt an in-process repair from the agent. Full steps are in the shared protocol's "Recovering a corrupt index" section.

• Conflicting facts. Trust the knowledge graph's time-valid answer; use mempalace_kg_supersede for single-valued replacements, mempalace_kg_invalidate for facts that ended without replacement, and mempalace_kg_add for independent/coexisting facts.

Anti-patterns — never do these

• Answering about past work, people, or decisions from model memory when the palace might know — search first.
• Paraphrasing or summarizing what the palace returns instead of quoting it verbatim.
• Searching on every turn, including greenfield tasks with no memory relevance.
• Pasting the whole conversation or a system prompt into the query argument — keep queries short and keyword-driven.

Official References

• MemPalace: https://github.com/MemPalace/mempalace
• MemPalace releases: https://github.com/MemPalace/mempalace/releases
• Cursor Skills documentation: https://cursor.com/docs/skills
• Agent Skills specification: https://agentskills.io/specification