[PR #808] [CLOSED] feat: add Aegis DQ MCP Agent — agentic data quality validation #6324

New Issue

GiteaMirror · 2026-05-27T23:09:43-05:00

GiteaMirror commented

2026-05-27 23:09:43 -05:00

📋 Pull Request Information

Original PR: https://github.com/Shubhamsaboo/awesome-llm-apps/pull/808
Author: @Shivakoreddi
Created: 5/14/2026
Status: ❌ Closed

Base: main ← Head: add-aegis-dq-mcp-agent

📝 Commits (2)

c90b823 feat: add Aegis DQ MCP Agent demo
593ad25 fix: fresh-start DB and add main block to setup_demo_db

📊 Changes

8 files changed (+380 additions, -0 deletions)

View changed files

📝 README.md (+1 -0)
➕ mcp_ai_agents/aegis_dq_mcp_agent/README.md (+104 -0)
➕ mcp_ai_agents/aegis_dq_mcp_agent/main.py (+166 -0)
➕ mcp_ai_agents/aegis_dq_mcp_agent/mcp_agent.config.yaml (+20 -0)
➕ mcp_ai_agents/aegis_dq_mcp_agent/mcp_agent.secrets.yaml.example (+2 -0)
➕ mcp_ai_agents/aegis_dq_mcp_agent/requirements.txt (+6 -0)
➕ mcp_ai_agents/aegis_dq_mcp_agent/sample_rules.yaml (+41 -0)
➕ mcp_ai_agents/aegis_dq_mcp_agent/setup_demo_db.py (+40 -0)

📄 Description

What

Adds a new MCP AI Agent demo: Aegis DQ MCP Agent (mcp_ai_agents/aegis_dq_mcp_agent/).

About

Aegis DQ is an open-source agentic data quality framework. This demo connects to it as an MCP server and lets users validate data, diagnose failures, and explore the audit trail — all through natural language.

What the demo does

Seeds a local DuckDB orders table with 5 intentional data quality issues (null IDs, negative amounts, invalid status, duplicates)
Connects to the Aegis DQ MCP server via mcp-agent
Users ask questions in natural language — the agent calls run_validation, list_runs, search_decisions, compare_reports
Results are shown as structured markdown with severity grouping

Files

mcp_ai_agents/aegis_dq_mcp_agent/
├── main.py                        # Streamlit app
├── setup_demo_db.py               # Seeds DuckDB with sample data
├── sample_rules.yaml              # 5 data quality rules to validate
├── mcp_agent.config.yaml          # MCP server config
├── mcp_agent.secrets.yaml.example # API key template
├── requirements.txt               # Dependencies
└── README.md                      # Setup + usage guide

How to run

cd mcp_ai_agents/aegis_dq_mcp_agent
pip install -r requirements.txt
export OPENAI_API_KEY=your-key
streamlit run main.py

Example prompts

"Run the rules at sample_rules.yaml against DuckDB with no_llm=true"
"Show me the last 5 validation runs"
"Search the audit trail for null customer IDs"

_{🔄 This issue represents a GitHub Pull Request. It cannot be merged through Gitea due to API limitations.}

## 📋 Pull Request Information **Original PR:** https://github.com/Shubhamsaboo/awesome-llm-apps/pull/808 **Author:** [@Shivakoreddi](https://github.com/Shivakoreddi) **Created:** 5/14/2026 **Status:** ❌ Closed **Base:** `main` ← **Head:** `add-aegis-dq-mcp-agent` --- ### 📝 Commits (2) - [`c90b823`](https://github.com/Shubhamsaboo/awesome-llm-apps/commit/c90b823fa76d84cd6e455ff10f29f36af86b8bf4) feat: add Aegis DQ MCP Agent demo - [`593ad25`](https://github.com/Shubhamsaboo/awesome-llm-apps/commit/593ad257d84b3e9ec22911bbee53dbcbb875e851) fix: fresh-start DB and add __main__ block to setup_demo_db ### 📊 Changes **8 files changed** (+380 additions, -0 deletions) <details> <summary>View changed files</summary> 📝 `README.md` (+1 -0) ➕ `mcp_ai_agents/aegis_dq_mcp_agent/README.md` (+104 -0) ➕ `mcp_ai_agents/aegis_dq_mcp_agent/main.py` (+166 -0) ➕ `mcp_ai_agents/aegis_dq_mcp_agent/mcp_agent.config.yaml` (+20 -0) ➕ `mcp_ai_agents/aegis_dq_mcp_agent/mcp_agent.secrets.yaml.example` (+2 -0) ➕ `mcp_ai_agents/aegis_dq_mcp_agent/requirements.txt` (+6 -0) ➕ `mcp_ai_agents/aegis_dq_mcp_agent/sample_rules.yaml` (+41 -0) ➕ `mcp_ai_agents/aegis_dq_mcp_agent/setup_demo_db.py` (+40 -0) </details> ### 📄 Description ## What Adds a new MCP AI Agent demo: **Aegis DQ MCP Agent** (`mcp_ai_agents/aegis_dq_mcp_agent/`). ## About [Aegis DQ](https://github.com/aegis-dq/aegis-dq) is an open-source agentic data quality framework. This demo connects to it as an MCP server and lets users validate data, diagnose failures, and explore the audit trail — all through natural language. ## What the demo does - Seeds a local DuckDB `orders` table with **5 intentional data quality issues** (null IDs, negative amounts, invalid status, duplicates) - Connects to the Aegis DQ MCP server via `mcp-agent` - Users ask questions in natural language — the agent calls `run_validation`, `list_runs`, `search_decisions`, `compare_reports` - Results are shown as structured markdown with severity grouping ## Files ``` mcp_ai_agents/aegis_dq_mcp_agent/ ├── main.py # Streamlit app ├── setup_demo_db.py # Seeds DuckDB with sample data ├── sample_rules.yaml # 5 data quality rules to validate ├── mcp_agent.config.yaml # MCP server config ├── mcp_agent.secrets.yaml.example # API key template ├── requirements.txt # Dependencies └── README.md # Setup + usage guide ``` ## How to run ```bash cd mcp_ai_agents/aegis_dq_mcp_agent pip install -r requirements.txt export OPENAI_API_KEY=your-key streamlit run main.py ``` ## Example prompts - "Run the rules at sample_rules.yaml against DuckDB with no_llm=true" - "Show me the last 5 validation runs" - "Search the audit trail for null customer IDs" --- <sub>🔄 This issue represents a GitHub Pull Request. It cannot be merged through Gitea due to API limitations.</sub>

GiteaMirror added the pull-request label 2026-05-27 23:09:43 -05:00

GiteaMirror closed this issue

2026-05-27 23:09:45 -05:00

Sign in to join this conversation.

1 Participants

Notifications

Due Date

No due date set.

Dependencies

No dependencies set.

Reference: github-starred/awesome-llm-apps#6324