tool-use

Select and sequence MCP tool calls with error handling.

Updated Aug 27, 2026
One-click install
npx skills add https://github.com/cris-m/flopsy --skill tool-use-cris-m
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: tool-use
Source: https://github.com/cris-m/flopsy/tree/main/src/team/templates/skills/tool-use
Command: npx skills add https://github.com/cris-m/flopsy --skill tool-use-cris-m

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Pattern-driven guidance for selecting, sequencing, and troubleshooting MCP tool calls to enable reliable automated workflows.

Core Features & Use Cases

  • Tool Selection Framework: Identify action, map to the right tool, gather parameters, and avoid unnecessary hops.
  • Tool Chaining Patterns: Sequential, parallel, and fallback patterns to compose tool calls safely.
  • Error Handling: Structured handling for authentication, permission, rate limits, Not Found, and validation errors.
  • Channel Messaging Discipline: Defines how status and results are communicated through channel-aware messaging.

Quick Start

Describe the user goal, map it to the most direct tool, and collect the required parameters before starting the first call.

Frequently Asked Questions about tool-use

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

FAQPage Schema
How do I handle MCP tool errors for authentication and rate limits?

Sequential, parallel, and fallback patterns enable safe tool chaining by orchestrating deterministic sequences of MCP tool calls, ensuring predictable outcomes when composing multiple automated workflows.

How do I handle MCP tool errors for authentication and rate limits?

Structured error handling addresses authentication, permission, rate limit, Not Found, and validation errors during MCP tool use, ensuring reliable automated workflows can recover safely from integration failures.

What is the best way to chain multiple MCP tool calls together?

Sequential, parallel, and fallback patterns allow you to safely chain MCP tool calls by orchestrating deterministic sequences, ensuring predictable outcomes when composing multi-tool automated workflows.

How do I troubleshoot MCP tool results that return unexpected outputs?

Troubleshoot unexpected MCP tool results by verifying collected parameters, applying structured error handling for validation failures, and checking the deterministic sequencing logic used during tool selection.

Do I need to collect all parameters before starting an MCP tool workflow?

Yes, you must collect clear parameters before starting the first MCP tool call to ensure deterministic tool sequencing, avoid unnecessary workflow hops, and guarantee predictable automation outcomes.