skills.rest
RamVegiraju
RamVegirajuCOMMUNITY
Explore all skills by this author →
WebsiteGitHub

modify-agent

Modify Databricks agent code and configurations for tailored behavior.

4|4|Updated Jan 5, 2026
Search Tags:#agent#mcp#openai#modify#tools#mlflow#databricks
One-click install
npx skills add https://github.com/RamVegiraju/databricks-samples --skill modify-agent
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: modify-agent
Source: https://github.com/RamVegiraju/databricks-samples/tree/main/.claude/skills/modify-openai-agent
Command: npx skills add https://github.com/RamVegiraju/databricks-samples --skill modify-agent

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Enables developers to modify agent code, add tools, or adjust configurations to tailor agent behavior for new tasks and environments.

Core Features & Use Cases

  • Modify core files (e.g., agent_server/agent.py) to change instructions, models, or MCP server configurations.
  • Add or update MCP servers and tool integrations via code changes and databricks.yml configuration.
  • Run, test, and validate agent changes locally using start_server.py and evaluate_agent.py, including MLflow-based tracing for diagnosability. Real-world example: update an agent to support a new MCP tool and switch to a more capable model.

Quick Start

Edit agent_server/agent.py to adjust agent behavior or add MCP servers, then restart the server.

Frequently Asked Questions about modify-agent

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I modify an agent's behavior and add MCP tools in a Databricks environment?▼

To modify agent behavior, edit the agent_server/agent.py file to adjust instructions, switch models, or add MCP tool integrations. Update databricks.yml configurations to register MCP servers, then restart start_server.py to apply changes locally.

Can I switch the OpenAI model used by my agent for better performance?▼

Yes, you can switch the OpenAI model by modifying the agent framework's code in agent.py. Adjust the model configurations to point to a more capable model, then validate the behavior changes using local tests and MLflow tracing.

What's the best way to test and evaluate agent code changes before deployment?▼

The best way to test agent changes is running evaluate_agent.py locally to validate behavior updates. Use MLflow-based tracing to diagnose issues and verify that new MCP tool integrations or model switches perform as expected before deployment.

Does this agent modification workflow support MCP server integrations?▼

Yes, the workflow fully supports MCP server integrations. You can add or update MCP servers by modifying code in agent.py and updating databricks.yml configurations to define the tool integrations across development and testing phases.

Why do I need MLflow tracing when modifying agent configurations?▼

MLflow tracing is needed to diagnose and evaluate agent behavior during modifications. It provides observability into how instruction changes, model switches, or new MCP tools affect agent performance, ensuring functional requirements are met before deployment.

What files do I need to edit to update Databricks agent instructions and tools?▼

You need to edit agent_server/agent.py to change instructions or models, and databricks.yml to configure MCP servers. Additionally, use start_server.py and evaluate_agent.py to run and validate your agent modifications locally.

Related Skills

Explore More Software Engineering Skills→
victorlouvictorlou
4

modify-agent

Modify LangGraph agent code, MCP servers, and Databricks permissions.

Community
Intermediate
RamVegirajuRamVegiraju
4

agent-evaluation

Evaluate GenAI agents with MLflow datasets, scorers, and traces.

Community
Advanced
databricksdatabricks
183

modify-agent

Modify agent code, integrate MCP servers, and update model configurations.

Official
Intermediate
sumitsaxena-gitsumitsaxena-git

modify-agent

Modify agent code, tools, and configuration for Databricks OpenAI Agents SDK workflows.

Community
Advanced