The agent that remembers everything.
Five-tier persistent memory with 96.6% recall on LongMemEval.
Every session starts from zero. Agents can't recall what you discussed yesterday, last week, or last month.
You re-explain your preferences, project decisions, and architectural choices in every single conversation.
Gotchas, debugging insights, and hard-won knowledge vanish when the session ends. Future sessions repeat past mistakes.
An agent without memory is a brilliant colleague with permanent anterograde amnesia — every conversation is their first day on the job.
Verbatim storage with embedding-based similarity search. 96.6% R@5 retrieval.
Subject-predicate-object triples. Cross-wing traversal and relationship queries.
HANDOFF.md, GLOSSARY.md, PROJECT_CONTEXT.md — structured markdown that survives git clones.
FTS5 full-text search for explicit key-value memories. Fast structured lookups.
Per-agent session journals. What was worked on, what was learned, what comes next.
Each tier excels at a different retrieval pattern. The palace metaphor — wings, rooms, drawers — gives agents an intuitive spatial navigation model across all five.
hooks-mempalace-briefing assembles a wake-up briefing from semantic search, knowledge graph, agent diary, and HANDOFF.md. Importance re-ranking surfaces what matters most.
hooks-mempalace-capture files verbatim tool outputs as palace drawers with auto-detected wing and room. No manual "remember this" required.
hooks-mempalace-interject surfaces relevant memories at the right moment. Only fires when cosine similarity ≥ 0.72 — non-disruptive by design.
The Curator agent runs three phases: palace curation, coordination file updates (HANDOFF.md, PROVENANCE.md), and KG enrichment with importance scoring and duplicate linking.
Recall@5 on LongMemEval — raw semantic retrieval, no LLM post-processing.
docs/research/gene-transfer-v1.2.0.pdf.
The interject hook surfaces relevant memories mid-session — but only when it's genuinely useful.
Fires on prompt_submit, tool_pre, and orchestrator_complete events. Requires cosine similarity ≥ 0.72 before surfacing anything. In the uncertain band, an LLM judge decides relevance.
Every hook emits structured events to ~/.mempalace/events/{session_id}.jsonl. Query with palace events or tail -f for live observability. Kill switch per hook: emit_events: false.
The worst memory system is the one that interrupts you with irrelevant trivia. MemPalace's interject hook stays silent unless it has something genuinely worth saying.
The read path. Precise palace search, KG queries, graph traversal, and coordination file reads. Invoked on-demand when you need to find something.
Conversational memory Q&A. Ask natural-language questions: "What decisions have I made about auth?" or "Summarize last week's work."
The write path. Palace curation, Phase 3 KG enrichment with importance scoring, HANDOFF.md updates, and duplicate linking. Runs at session end and on-demand.
On-demand deep structural analysis: BFS clustering, KG edges, diary entries, and importance backfill. Zero deletion — all outputs are additive KG facts, capped at 500 drawers with a 120s timeout.
Coordination files (PROJECT_CONTEXT.md, GLOSSARY.md, HANDOFF.md) are readable by Amplifier, OpenAI Codex, GitHub Copilot, Cursor, and Windsurf. Memory that travels with the repo.
4,313 lines of source, 5,347 lines of tests across 5 modules. Integration tests exercise the full palace lifecycle — briefing, capture, interject, and curation.
tool-mempalace — 8 operations: search, remember, kg, traverse, diary, mine, events, gardenhooks-mempalace-briefing — session-start wake-uphooks-mempalace-capture — verbatim auto-filinghooks-mempalace-interject — mid-session surfacinghooks-project-context — coordination filesAll data stays on your machine. Palace storage at ~/.mempalace/. Nothing leaves your environment.
Duplicates are preserved with low importance, never dropped. KG enrichment and garden analysis are purely additive.
Without MemPalace installed, the bundle runs in coordination-files-only mode. Palace features silently skip.
MemPalace consolidates five previously separate memory modules into a single bundle. It's the persistence layer that makes agent continuity possible.
Supersedes amplifier-bundle-memory, amplifier-bundle-project-memory, amplifier-module-context-memory, amplifier-module-tool-memory, and amplifier-module-hooks-memory-capture into one cohesive system.
Includes a 14-page research paper covering architecture, event observability design, KG intelligence with formula proofs, evaluation methodology, and philosophy preservation analysis.
The palace metaphor isn't decorative — it gives agents a spatial navigation model. Wings are projects, rooms are topics, drawers are memories. The same mental model humans have used since ancient Greece.
Primary repository: colombod/amplifier-bundle-memory
Data sources:
git log, git shortlog -sne, find | wc -l on cloned repocapabilities/colombod/amplifier-bundle-memory/memory.yamlBenchmark data:
docs/research/gene-transfer-v1.2.0.pdfNo metrics were fabricated. All numbers derived from git history and repository documentation as of May 2026.
Install MemPalace and the memory bundle. Your agent will never forget again.