tool-use

tool-use is a skill for Claude Code, Codex from hajekim/agentic-design-patterns-extension. It costs 365 tokens per session (3,771 once invoked), scanned A, a copy of tool-use, MIT.

A pattern for letting an AI agent use outside tools such as web searches, APIs, databases, code runners, and file or memory stores. The agent chooses a tool, sends it inputs, and uses the result in its answer.

In plain words
What is it for?
Use it to add web search, document retrieval, database queries, code execution, persistent storage, email, calendar events, or other external services to an agent.
Why use it?
It handles tasks that need current information, calculations, stored data, or real-world actions that the language model cannot do by itself.

Skill for Claude CodeCodex

Written for no agent in particular: nothing here depends on one.

Good fit Use it to add web search, document retrieval, database queries, code execution…

Compare 6 skills from other repositories ↓
Install with agentmods
npx agentmods add skills/hajekim/agentic-design-patterns-extension/tool-use
Install

Getting it into your agent

One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.

Any agent
npx skills add hajekim/agentic-design-patterns-extension --skill tool-use
Clone the repo
git clone --depth 1 https://github.com/hajekim/agentic-design-patterns-extension

Made for: Claude Code, Codex.

Wrote this? Show the measurements

A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.

agentmods badge for tool-use

README.md
[![agentmods](https://agentmods.dev/badge/skills/hajekim/agentic-design-patterns-extension/tool-use.svg)](https://agentmods.dev/skills/hajekim/agentic-design-patterns-extension/tool-use)
Your own site
<a href="https://agentmods.dev/skills/hajekim/agentic-design-patterns-extension/tool-use"><img src="https://agentmods.dev/badge/skills/hajekim/agentic-design-patterns-extension/tool-use.svg" alt="Measured on agentmods" height="20"></a>
Per session 365 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 3,771 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 2 findings. A grade says what 26 rules found in the file — not that it is safe.
Origin 100% copy Near-identical to another mod in the catalogue.
Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

ModelPer sessionOnce invoked
Fable 5.1 $0.00365 $0.03771
Opus 5 $0.00182 $0.01886
Sonnet 5 $0.00073 $0.00754
Haiku 4.5 $0.00036 $0.00377

Measured 6d ago against content hash 5c17edbbe6e7, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-06, from the pricing page.

Security

Grade A, and why

tool-use scanned grade A with 2 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 6d ago.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Makes network callslowCapability

Not a fault in itself. Listed so you know the mod talks to something, and to what.

response = requests.get(endpoint, params=params, timeout=10)

Runs shell commandslowCapability

Expected in a hook, worth knowing in a rule or an instructions file.

result = subprocess.run(
Origin

This is a copy

100% identical to tool-use — 3 lines differ, which has more behind it and is treated as the original. This page carries a canonical link to it rather than competing with it.

skills/tool-use/SKILL.md · 438 lines

How it starts

The opening of the file, as written. The whole thing — 438 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Tool Use Pattern

Overview

The Tool Use Pattern extends an agent's capabilities beyond pure language generation by enabling it to interact with external systems — APIs, databases, search engines, code executors, and more. The agent reasons about which tool to call, invokes it with appropriate parameters, processes the returned data, and integrates the results into its response.

Core Principle: Extend the agent's reach — let it act in the world, not just reason about it.

When This Skill Applies

Activate this pattern when:

  • The agent needs real-time or external data (weather, stock prices, news)
  • Tasks require computation, code execution, or mathematical verification
  • Document retrieval and RAG (Retrieval-Augmented Generation) is needed
  • The agent must persist state (write to databases, files, memory stores)
  • Actions must be taken in the world (send email, create calendar events, trigger workflows)
  • The agent needs to verify its reasoning with factual lookups

Rule of thumb: If the task requires information or capabilities beyond the model's training data or inherent abilities — use tools.

DEFINE → PLAN → ACTION Workflow

DEFINE

Map the tool requirements:

  1. What external data or capabilities does the agent need?
  2. What tools exist or need to be built to provide those capabilities?
  3. What are the input/output schemas for each tool?
  4. What are the failure modes and how should they be handled?

PLAN

Design the tool integration:

  1. Define each tool with a clear name, description, and parameter schema
  2. Write tool descriptions that help the LLM understand when and how to use each tool
  3. Design the tool selection logic (LLM-driven function calling vs. rule-based)
  4. Plan error handling for API failures, timeouts, and unexpected responses
  5. Consider tool chaining — sequential tool calls where outputs feed subsequent calls

ACTION

Implement tool-enabled agents:

  1. Register tools with the LLM using the framework's function calling interface
  2. Implement each tool function with proper input validation and error handling
  3. Let the LLM determine when and how to call tools based on user intent
  4. Process tool results and integrate them into the agent's response
  5. Log tool calls for observability and debugging

Read the full file on GitHub · 438 lines

Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

  1. 6d ago First seen · 438 lines · 365 tokens per session scan A 5c17edbbe6e7

Subscribe to this mod's changes

tool-use is a skill published in the GitHub repository hajekim/agentic-design-patterns-extension (1 stars, last pushed 5mo ago), licensed MIT. It adds 365 tokens to every session and 3,771 once invoked, about $0.0018 per session on Opus 5. A static security scan graded it A with 2 findings (makes network calls, runs shell commands). It is 100% identical to tool-use, differing in 3 lines, and is treated as a copy.