Stxyu-p

Memcore — Data skill for Claude Code

Data community

Global, governed, bitemporal memory engine for autonomous AI coding agents & IDEs — shared ~/.memcore brain across Claude Code, Cursor, Antigravity, Codex, Gemini & Hermes with SQLite WAL + FTS5 + sql.

How to install Memcore

This entry records only its repository, not the path inside it, so there is no exact command to give. Open Stxyu-p/memcore and copy the folder into ~/.claude/skills/, or the file into ~/.claude/agents/.

What Memcore does

Global, governed, bitemporal memory engine for autonomous AI coding agents & IDEs — shared ~/.memcore brain across Claude Code, Cursor, Antigravity, Codex, Gemini & Hermes with SQLite WAL + FTS5 + sqlite-vec SIMD + model2vec hybrid RRF recall over stdio MCP.

Alternatives in Data

  • Azure Cosmos DB .net — Cosmos DB NoSQL with global distribution 1.8k ★
  • Autonovel — An autonomous novel writing pipeline, by Hermes Agent 1.5k ★
  • Global Stock Data — US stock market data for AI coding assistants — zero-auth, official sources 1.5k ★

README

MemCore

**Governed, Bitemporal, Local-First Memory Engine for Autonomous AI Agent Fleets**

Capture broadly. Recall narrowly. Never trust raw conversational history.

[![PyPI](https://img.shields.io/pypi/v/memcore-omnix?style=flat-square&color=blue)](https://pypi.org/project/memcore-omnix/) [![Python 3.10+](https://img.shields.io/badge/python-3.10+-blue?style=flat-square)](https://www.python.org/) [![Storage](https://img.shields.io/badge/storage-SQLite_WAL_+_FTS5-07405E?style=flat-square)](#why-memcore) [![Latency](https://img.shields.io/badge/latency-%3C_3ms_hybrid_recall-10b981?style=flat-square)](#sub-millisecond-hybrid-rrf-recall) [![Governance](https://img.shields.io/badge/governance-Bitemporal_+_Tombstones-8b5cf6?style=flat-square)](#the-four-pillars-of-governed-memory) [![License: MIT](https://img.shields.io/badge/license-MIT-slate?style=flat-square)](LICENSE)


The Problem: State Poisoning in Ungoverned Vector Memory

Most agent memory frameworks dump raw conversational logs directly into uncurated vector databases. Over multi-session projects, this inevitably creates three compounding failures:

  1. State Poisoning & Compounding Errors: When an agent makes an incorrect guess once, that wrong guess is embedded and recalled as "truth" in all future sessions.
  2. Stale Code Drift: Facts about an architecture stay frozen in memory long after the underlying files have been refactored or deleted.
  3. Cross-Agent Isolation Leaks: Private agent scratchpads bleed into shared team context without hard security boundaries.

The Contrast: Ungoverned Vector RAG vs. MemCore

**Ungoverned Vector Memory:**

Query: "What port does the analytics service use?"
Recalled Chunks (Unfiltered Cosine Similarity):
1. [0.89] "Let's try port 8123..." (Old failed guess)
2. [0.87] "Runs on port 9000 in settings.py" (Stale config from last week)
3. [0.85] "Migrated to port 9440 with mTLS" (Actual current truth)
Result: ❌ Contradictory history dumped into prom