ACI
Verified Safeby az9713
Overview
A headless MCP (Model Context Protocol) server that enables AI agents like Claude to read, write, and reason over a knowledge graph of scientific propositions.
Installation
uv run python src/server.pyEnvironment Variables
- OPENAI_API_KEY
Security Notes
The system explicitly states 'No authentication - Designed for local/personal use'. This means anyone with local access can perform all operations, including deletion. There is a potential for injection vulnerabilities in LanceDB's `where` and `delete` clauses (e.g., `table.search().where(f"id = '{unit_id}'")`) if `unit_id` can be controlled by a malicious input and is not strictly validated (e.g., as a UUID format). While `unit_id` is often internally generated, the `get_unit` and `delete_unit` tools accept arbitrary strings as `unit_id` without explicit UUID format validation before constructing the f-string query. This could be exploited by a sophisticated or compromised agent. The `delete_unit` tool has a `confirm` flag, but a malicious agent could bypass this by providing `confirm=True`.
Similar Servers
gpt-researcher
The GPT Researcher MCP Server enables AI assistants to conduct comprehensive web research and generate detailed, factual, and unbiased reports. It supports multi-agent workflows, local document analysis, and integration with external tools via the Machine Conversation Protocol (MCP) for various research tasks.
claude-flow
AI Agent Orchestration and Development Platform for Claude Code
zen-mcp-server
A server for coordinating and managing AI agents, likely for simulations or complex task execution, leveraging Claude LLMs.
Google-Search-MCP-Server
This MCP server enhances Google search with AI-powered research synthesis, content extraction, source quality assessment, and deduplication, designed to be used by large language models (LLMs) like Claude.