1. Conduid
  2. RAG
  3. Loanguard AI
MCP server · RAG

Loanguard AI

· github.com/emillaurence/loanguard-ai

Agentic GraphRAG system for loan compliance monitoring and financial crime investigation in Australian financial services. Built on Neo4j, Claude, and APRA prudential standards. Features a multi-agent pipeline, three-layer knowledge graph architecture, and full reasoning chains with cited regulatory evidence.

Unclaimed agentic-aiagentsaiclaudeknowledge-graphneo4j
InstallView source
39Low

Scored a month ago · breakdown

About Loanguard AI

Loanguard AI is an MCP server in the RAG category: agentic GraphRAG system for loan compliance monitoring and financial crime investigation in Australian financial services. Built on Neo4j, Claude, and APRA prudential standards. Features a multi-agent pipeline, three-layer knowledge graph architecture, and full reasoning chains with cited regulatory evidence.

Install

From source
git clone https://github.com/emillaurence/loanguard-ai

This server has no ConduID identity, so agent calls to it are not receipted. Pin the version you install and review the source before granting it credentials.

Tools (6)

detect_graph_anomaliesRuns one or more named anomaly detection patterns from `ANOMALY_REGISTRY`. Accepts a list of pattern names and an optional `entity_id` to scope results.

Parameters not published.

evaluate_thresholdsEvaluates a list of `Threshold` nodes against an entity's stored property values. Uses `threshold_type` to determine whether each threshold results in PASS, BREACH, TRIGGER, or N/A, and filters out `informational` thresholds from verdict logic.

Parameters not published.

persist_assessmentIdempotent MERGE to Layer 3. Creates `Assessment`, `Finding`, and `ReasoningStep` nodes linked to the entity and regulation. Returns the `assessment_id`.

Parameters not published.

retrieve_regulatory_chunksSemantic search over `Chunk` nodes using OpenAI `text-embedding-3-small` embeddings and cosine similarity. Accepts an optional `regulation_id` filter and returns top-k chunks with similarity scores.

Parameters not published.

trace_evidenceWalks a stored `Assessment` back through its `ReasoningStep` nodes to retrieve all cited `Section` and `Chunk` nodes. Used by the Orchestrator to populate `cited_sections` and `cited_chunks`.

Parameters not published.

traverse_compliance_pathCross-layer L1 to L2 traversal via the Jurisdiction bridge. Returns the full regulatory subgraph for an entity, including all thresholds with their `threshold_type` field.

Parameters not published.

Ask AI

✦

Ask AI about Loanguard AI

Powered by Claude · Grounded in docs

I know everything about Loanguard AI. Ask me about installation, configuration, usage, or troubleshooting.

Security checks

6 checks have not been run yet.

Questions

About Loanguard AI

How do I install Loanguard AI?

No package has been published for this server, so install it from source: clone the repository and follow its README.

Is Loanguard AI safe to use with an AI agent?

Its trust score is 39 out of 100 (low). Conduid hasn't run static security checks on this repository yet, so review the source yourself before granting it credentials. It has no ConduID identity yet, so agent calls to it are not receipted.

Is Loanguard AI still maintained?

Conduid hasn't recorded a commit date for this repository yet. Check the repository directly for recent activity.