arthurpanhku/Arthor-Agent

Overview

Arthor-Agent is an MCP server for an AI agent that automates cybersecurity document assessment. It parses multiple formats and uses a RAG knowledge base to identify risks, compliance gaps, and remediations in questionnaires and reports.

Best for

Best for
Security teams automating document review and compliance checks

Use cases

• Automate security questionnaire responses
• Assess compliance documents for gaps
• Extract risks and remediations from reports

How to use

Install

pip install -r requirements.txt

Tools exposed

• LLM_PROVIDER
• EVIDENCE_CRITIC_ENABLED
• EVIDENCE_CRITIC_MAX_CANDIDATES
• EVIDENCE_CRITIC_MAX_THREATS
• CHROMA_PERSIST_DIR
• PARSER_ENGINE
• ENABLE_GRAPH_RAG
• LANGGRAPH_CHECKPOINT_DIR
• SSDLC_DEFAULT_PHASES
• SSDLC_DEFAULT_STAGE
• MCP_DOCUMENT_ROOTS
• AGENT_GATEWAY_ENABLED
• AGENT_GATEWAY_TOKEN
• AGENT_GATEWAY_TENANT_ID
• AGENT_GATEWAY_PUBLIC_URL
• AGENT_GATEWAY_ALLOWED_HOSTS
• AGENT_GATEWAY_ALLOWED_ORIGINS

Tested with

Claude Desktop, Cursor, ChatGPT

Notes

Arthor-Agent is an MCP server for an AI agent that automates cybersecurity document assessment. It parses multiple formats and uses a RAG knowledge base to identify risks, compliance gaps, and remediations in questionnaires and reports.

90 stars on GitHub. Last updated 2026-05-28. Licensed MIT.

Use cases

• Automate security questionnaire responses
• Assess compliance documents for gaps
• Extract risks and remediations from reports

Pros

• Open source Python implementation on GitHub
• Multi-format parsing with RAG for context-aware analysis
• Directly addresses compliance and risk assessment workflows

Cons

• Low star count (90) indicates limited community adoption
• Requires setup of MCP server and knowledge base
• May need customization for specific regulatory frameworks

Indexed from awesome-mcp-servers-punkpeye and enriched against its public facts.