Agentic Coding & AI Dev Tools5.0 · 0 ratings

Agent Tool Selection Decision Guide

Helps an agent pick the right tool among overlapping options and avoid wrong or redundant calls.

Chain-of-ThoughtTree-of-Thoughts

Prompt

You are a Tool-Use Strategist who writes decision guidance so an agent picks the correct tool the first time among overlapping options.

Context: The agent has tools [TOOL_LIST], several with overlapping purposes. Observed mistakes: [OBSERVED_MISSELECTIONS]. The agent operates in [DOMAIN].

Reason step by step:
1. For each pair of overlapping tools, articulate the distinguishing criterion.
2. Build a selection decision tree keyed on the agent's intent.
3. Define anti-patterns: when NOT to use each tool.
4. Specify fallbacks when the preferred tool fails or is unavailable.
5. Add a quick lookup table mapping common intents to the right tool.

Output format:
### Overlap Disambiguation (table: tool A vs tool B | choose A when | choose B when)
### Selection Decision Tree
### Anti-Patterns Per Tool
### Fallback Chains
### Intent-To-Tool Lookup

Constraints: Criteria must be observable from the agent's intent, not hidden state. Every fallback must terminate. Prefer the cheapest tool that satisfies the intent. Do not recommend tools not in the list. Use [SQUARE_BRACKET] placeholders for domain specifics.

How to use this prompt

  1. 1

    Copy the prompt above and paste it into ChatGPT, Claude, or Gemini — or open it in the visual Studio to edit each part on a canvas and run it with your own key.

  2. 2

    Replace any bracketed placeholders with your specifics. The more concrete your context and constraints, the sharper the result — see the 5-part prompt structure.

  3. 3

    Run it, then refine. Ask the model to critique and improve its own answer with self-critique prompting.

Techniques in this prompt

Chain-of-Thought

Asks the model to reason step by step before answering — ideal for multi-step, logical, or analytical tasks.

Learn this technique
Tree-of-Thoughts

A tree of thoughts technique used to shape and strengthen the model's response.

Recommended models

claudegpt-4ogemini

Build on this prompt

Open it in the visual Studio to wire it into a full workflow with your own API key — or learn the craft behind prompts like this.

More in Agentic Coding & AI Dev Tools