gacerioni

Redis Agentcore Settle Agent — Data skill for Claude Code

Data community

Debt settlement agent for a fictional bank: Claude on Amazon Bedrock AgentCore Runtime with Redis Cloud as the context layer (semantic router, LangCache, Agent Memory, Context Retriever MCP tools, hyb.

How to install Redis Agentcore Settle Agent

This entry records only its repository, not the path inside it, so there is no exact command to give. Open gacerioni/redis-agentcore-settle-agent and copy the folder into ~/.claude/skills/, or the file into ~/.claude/agents/.

What Redis Agentcore Settle Agent does

Debt settlement agent for a fictional bank: Claude on Amazon Bedrock AgentCore Runtime with Redis Cloud as the context layer (semantic router, LangCache, Agent Memory, Context Retriever MCP tools, hybrid search)

Alternatives in Data

  • Private GPT — Complete API layer for private AI applications on local models: RAG, skills, tools, MCP, text-to-sql, and more 57.5k ★
  • 12 Services Layer — Prompt 12: Wire Up Services Layer (Analytics, Policy, Settings, Sessions) 2.3k ★
  • Sprite Gen — Generate clean 2D game sprites & animation atlases — component-row pipeline: state rows, alpha cleanup, frame 763 ★

README

Settle · a bank agent on Redis Cloud and Amazon Bedrock AgentCore

[![MIT License](https://img.shields.io/badge/license-MIT-green.svg)](LICENSE) [![Python 3.12](https://img.shields.io/badge/python-3.12-blue.svg)](pyproject.toml) [![Redis Cloud](https://img.shields.io/badge/Redis%20Cloud-LangCache%20%C2%B7%20Agent%20Memory%20%C2%B7%20Context%20Retriever-DC382D.svg)](https://redis.io/docs/latest/develop/ai/context-engine/) [![Amazon Bedrock AgentCore](https://img.shields.io/badge/Amazon%20Bedrock-AgentCore%20Runtime-FF9900.svg)](https://aws.amazon.com/bedrock/agentcore/)

Reference demo of a **debt settlement agent** for Gabs Bank (a fictional bank with fictional data). The agent is **Claude on Amazon Bedrock**, orchestrated with **LangChain and LangGraph**, runs on the **Amazon Bedrock AgentCore Runtime**, and uses **Redis Cloud** as its context layer: rate limiting and semantic routing before the model, semantic caching, short- and long-term memory, live business data as MCP tools, and hybrid search over versioned policies.

**The thesis in one sentence:** the model thinks; Redis remembers, filters, caches, and serves the current data. Every turn shows on screen, in three blocks, what happened before the model, what the agent called, and the result.

What this repository shows

Capability Redis Cloud service In the code
Per-customer rate limit and semantic routing with guardrails (off-topic and prompt injection die before the model; the route picks Sonnet or Haiku) Redis (Search, Lua) settle/ratelimit.py, settle/router.py
Semantic cache scoped by customer segment and knowledge-base version, with selective invalidation and no flush Redis LangCache with attributes settle/cache.py
Session memory and long-term memory per customer, with automatic extraction Redis Agent Memory settle/memory.py
Live customer data (contracts, payment profile, agreements) as MCP tools generated from the data model **Redis