Happy slice of mcp pizzaMCP.pizza

world-model-mcp

MCP.Pizza Chef: SaravananJaichandar

This server adds a lasting memory and secure audit trail for AI coding assistants like Claude Code, Cursor, and VS Code extensions. It helps your AI avoid repeating errors, makes hallucinations verifiable, and tracks corrections over time. It runs locally, requires developer setup, and does not need online accounts or keys. Use it to improve reliability and trust in AI-assisted coding workflows.

Coding
Other

Use This MCP server To

Keep AI coding assistants from making the same mistakes twice Verify AI-generated code to reduce hallucinations Track changes and corrections across coding sessions Audit AI coding agent actions with cryptographic proof Use with Claude Desktop or VS Code for smarter coding help Maintain a local knowledge graph for AI coding context

README

World Model MCP

Persistent memory + optional signed audit for AI coding agents.

world-model-mcp ships a local SQLite knowledge graph your agent queries every turn: hallucinations become verifiable, corrections stick across sessions, and regressions get caught before they land. Flip on the audit chain and every event is signed with FIPS 205 hybrid Ed25519 + SLH-DSA, verifiable offline forever. MIT-licensed, runs entirely local, works with 10+ AI coding agents including Claude Code, Cursor, Codex, Continue, Cline, Windsurf, GitHub Copilot Chat, pi, OpenClaw, and Hermes Agent.

Latest: v0.15.8. Docs-only refresh. No code changes. Adds a "Which one is right for me?" guide in the README to help a first-time reader decide whether the local world-model-mcp server is enough or whether they also want the hosted etch.systems companion for signed audit evidence. Every code path shipped in v0.15.7 continues to work identically. Full version history in CHANGELOG.md.

PyPI Downloads License: MIT Python 3.11+ world-model-mcp MCP server DOI

mcp-name: io.github.SaravananJaichandar/world-model-mcp

Compared to other agent-memory + agent-audit projects

Head-to-head positioning against eight named peers on the audit / signing / anchoring dimensions the agent-memory space is converging on. Every world-model-mcp cell carries a provenance flag: own = measured / observed in the shipped product; cited = pulled from the competitor's own public landing, repo, or press.

Source of truth for this table lives in the hosted service repo: world-model-mcp-hosted/src/etch/competitor_matrix.py. Update in both places if you change either.

world-model-mcp FAQ

Can I use this to prevent AI coding agents from repeating mistakes?
Yes — it keeps a persistent memory that helps AI coding assistants avoid repeated errors across sessions.
Can I use this to verify AI-generated code for hallucinations?
Yes — it provides a knowledge graph and audit chain that makes hallucinations verifiable and corrections persistent.
Which apps work with world-model-mcp?
It supports Claude Desktop, Cursor, VS Code, Cline, Goose, Windsurf, and others.
Do I need an API key or account to use this?
No — it runs locally without requiring external accounts or keys.
How hard is it to set up?
It requires developer-level setup, including installing via pip or remote URL and some configuration.
Does it keep an audit trail of AI coding agent actions?
Yes — it signs every event with strong cryptography for offline verification.
Is this suitable for non-developers?
It’s mainly for developers or technical users due to setup complexity and integration needs.