skills.rest
leary-poken
leary-pokenCOMMUNITY
Explore all skills by this author →
GitHub

tool-selection

Evaluate agent traces to flag Bash workarounds and recommend Databricks MCP alternatives.

1|Updated Mar 17, 2026
Search Tags:#bash#evaluation#mcp#tool-selection#databricks#trace-analysis
One-click install
npx skills add https://github.com/leary-poken/ai-dev-kit --skill tool-selection-leary-poken
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: tool-selection
Source: https://github.com/leary-poken/ai-dev-kit/tree/main/.test/eval-criteria/tool-selection
Command: npx skills add https://github.com/leary-poken/ai-dev-kit --skill tool-selection-leary-poken

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

Evaluates whether agent tool usage favors Databricks MCP tools over shell workarounds, ensuring safer, faster, and repeatable automation.

Core Features & Use Cases

  • Enforces MCP-first tool usage rules for common operations (SQL, file and job management, workspace access).
  • Detects and reports Bash-based workarounds in traces, offering MCP-alternative recommendations.
  • Defines a clear evaluation rubric (MCP vs Bash, task-appropriate tool, error handling, and efficiency considerations) for production-ready workflows.

Quick Start

Ask the agent to run a trace evaluation and replace any Bash tool calls with the corresponding MCP tool calls when available.

Frequently Asked Questions about tool-selection

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

FAQPage Schema
How do I evaluate agent traces to enforce MCP tool usage over Bash workarounds?▼

To evaluate agent traces for MCP tool usage, apply a tool-selection rubric to your logs that flags Bash tool calls and recommends corresponding Databricks MCP alternatives for SQL, file, and workspace operations.

Why does my Databricks agent use Bash workarounds instead of the available MCP tools?▼

Agents often default to Bash workarounds when MCP tool mapping is not enforced. Evaluating your traces identifies these suboptimal choices and provides MCP alternatives to ensure safer, faster, and repeatable automation.

Can I detect and replace shell workarounds in my MCP tool traces?▼

Yes, you can detect shell workarounds by analyzing traces for Bash tool calls. The evaluation flags these inefficiencies and recommends replacing them with appropriate MCP tool calls for job management and workspace access.

What is the best way to map MCP and non-MCP tools for Databricks automation?▼

The best way to map MCP and non-MCP tools is to apply an evaluation rubric to your agent traces. This enforces task-appropriate tool usage for SQL and file operations, reporting deviations where Bash is used over MCP.

Does MCP tool evaluation support error handling and efficiency checks for Bash calls?▼

Yes, MCP tool evaluation includes error handling and efficiency considerations within its rubric. It assesses whether Bash tool calls in traces are necessary or if MCP alternatives provide better production-ready workflow outcomes.

When should I not use Bash workarounds for Databricks SQL and workspace operations?▼

You should avoid Bash workarounds when Databricks MCP tools are available for SQL, file, and workspace operations. Using Bash limits repeatability and safety, making trace evaluation necessary to flag and replace these suboptimal choices.

Related Skills

Explore More Software Engineering Skills→
itsadijmbtitsadijmbt
1

tool-selection

Evaluate Databricks agent traces for MCP tool selection over Bash workarounds.

Community
Intermediate
itsadijmbtitsadijmbt
1

sql-correctness

Validate Databricks SQL traces against Unity Catalog and safety rules.

Community
Intermediate
databricksdatabricks
183

add-tools

Configure Databricks agent access to external resources in databricks.yml.

Official
Intermediate
tanujsutariatanujsutaria
1

design-system-auditor

Audit SwiftUI views for hardcoded colors, spacing, and typography.

Community
Basic