Palace MCP server

palace-rs

A local-first memory retrieval engine for coding agents, implemented in Rust.

This project stores verbatim project and conversation memory, embeds it locally, and retrieves source-grounded context through MCP. It is built for coding agents that need to remember decisions, prior fixes, commands, project conventions, and user preferences across sessions without running a separate vector database.

What It Does

fuzzy "what do I prefer?" and convention questions.

• Stores project files and conversation turns in a local SQLite database.
• Generates local embeddings with ONNX Runtime and all-MiniLM-L6-v2.
• Retrieves memories with hybrid semantic/BM25 search plus coding-agent intent boosts.
• Tags preference-shaped drawers and runs a dedicated preference recall pass for

preference_match score for preference-shaped queries.

• Stores preference spans with optional secondary embeddings and exposes a

temporal, unknown) and can optionally rerank top results with a local interaction reranker.

• Classifies search intent (preference, decision, how_to, definition,

and compact session text.

• Sanitizes agent-generated query dumps before retrieval.
• Returns source-grounded results with score provenance and nearby source context.
• Warms up agents from recent diary entries with project, topic, timestamp, tags,

optional folded feedback on the existing palace_gain MCP tool.

• Provides a knowledge graph for temporal entity relationships.
• Measures real-world usefulness through palace gain precision metrics and

project mining and topic-wing creation.

• Exposes MCP tools for assistants that support Model Context Protocol.
• Offers a small Rust library API for embedding memory into other services.
• Tracks a first-class wings registry (project vs. topic wings) with on-demand

memory protocol doesn't depend on tool-search deferral (Claude Code >= 2.1.121).

• Pins the nine protocol-critical MCP tools resident with alwaysLoad so the

session's project — directly into SessionStart, not just protocol text.

• Injects real recalled memory — recent diary entries plus top drawers for the

Agent Memory Reliability

Palace focuses on the retrieval cases that matter most during coding work: preferences, project conventions, recent session continuity, source-grounded answers, and measurable usefulness in real agent sessions. Drawers that look like user preferences or conventions are tagged in metadata during writes and updates, record the matched preference span, and can store a secondary preference embedding. Preference-shaped queries receive a dedicated preference_match score alongside hybrid semantic/BM25 search.

MCP search responses expose score provenance (combined, cosine, bm25, and codingboost, preferencematch, optional rerank_score, and intent) plus adjacent source context so agents can cite why a memory was returned. Diary tools provide warm-start context for recent sessions, including project path, topic, timestamp, session ID, tags, and compact text.

Library consumers can use Palace::searchwithprovenance when they need the same structured score details that MCP tools return.

Storage

Collapses Python's dual-store (ChromaDB + SQLite) into one file at ~/.palace/palace.db:

Embeddings are stored as f32 vectors from all-MiniLM-L6-v2. Search uses local cosine similarity over the stored vectors.

Benchmarks

Coding-Agent Memory Eval

Shortened. The full README is on GitHub.