Ai Rete Rag MCP server
{
"mcpServers": {
"ai-rete-rag": {
"command": "uvx",
"args": [
"ai-rete-rag-mcp"
],
"env": {
"AI_RETE_RAG_API_KEY": "<ai_rete_rag_api_key>",
"AI_RETE_RAG_API_URL": "https://..."
}
}
}
}{
"mcpServers": {
"ai-rete-rag": {
"command": "uvx",
"args": [
"ai-rete-rag-mcp"
],
"env": {
"AI_RETE_RAG_API_KEY": "<ai_rete_rag_api_key>",
"AI_RETE_RAG_API_URL": "https://..."
}
}
}
}{
"mcpServers": {
"ai-rete-rag": {
"command": "uvx",
"args": [
"ai-rete-rag-mcp"
],
"env": {
"AI_RETE_RAG_API_KEY": "<ai_rete_rag_api_key>",
"AI_RETE_RAG_API_URL": "https://..."
}
}
}
}{
"servers": {
"ai-rete-rag": {
"command": "uvx",
"args": [
"ai-rete-rag-mcp"
],
"env": {
"AI_RETE_RAG_API_KEY": "${input:ai_rete_rag_api_key}",
"AI_RETE_RAG_API_URL": "${input:ai_rete_rag_api_url}"
}
}
},
"inputs": [
{
"id": "ai_rete_rag_api_key",
"type": "promptString",
"description": "Ai rete rag api key",
"password": true
},
{
"id": "ai_rete_rag_api_url",
"type": "promptString",
"description": "Ai rete rag api url",
"password": true
}
]
}{
"mcpServers": {
"ai-rete-rag": {
"command": "uvx",
"args": [
"ai-rete-rag-mcp"
],
"env": {
"AI_RETE_RAG_API_KEY": "<ai_rete_rag_api_key>",
"AI_RETE_RAG_API_URL": "https://..."
}
}
}
}file .mcp.json (in your project root)file claude_desktop_config.jsonfile ~/.cursor/mcp.jsonfile .vscode/mcp.jsonfile ~/.codeium/windsurf/mcp_config.jsoninstall uvx ai-rete-rag-mcpversion 0.7.0
What it does
This server makes deterministic rule-based decisions through a Rete engine while using an LLM to explain the reasoning grounded in policy documents. It is for developers who need auditable, reproducible decisions with human-readable justifications in domains like lending, fraud detection, or clinical triage.
- Evaluate a loan application against underwriting rules and receive a verdict with an explanation citing the relevant policy sections.
- Ingest a written policy document and automatically generate draft rules that encode each policy requirement, then review and deploy them.
- Inspect and audit a domain's rule set to understand conditions, verdicts, salience ordering, and rule overlaps.
Tools this server exposes
| Tool | Does |
|---|---|
| decide | Make a decision in a domain using structured facts or free text, with optional retrieval patterns to scope rules or populate working memory. |
| list_rules | Inspect a domain's rules including conditions, verdicts, salience, and overlaps. |
| get_rule_source | Fetch a domain's rule set as editable YAML. |
| import_policy_rules | Convert a written policy document into draft rules, each citing the source sentence, without saving. |
| put_rules | Create or replace a domain's rule set from YAML, with optional dry-run validation. |
| ingest_text | Add policy text to a domain's knowledge base for RAG-powered explanations. |
| list_documents | Browse the ingested policy documents in a domain. |
| get_usage | Check your plan and remaining monthly decision quota. |
Before you enable it
An MCP server runs with whatever access you give it. Start with a development account rather than production, and check which tools the server exposes before you let an agent call them unattended.