Loading...
Loading...
Loading...
3,528 documents
This is not a styling job. This is not a facelift.
description: Virtual office dashboard — pixel-art NPCs for your OpenClaw agents. Install, manage, and interact with your retro AI office.
> Track AI trading agents across Solana (pump.fun) and Monad (nad.fun) blockchains with real-time trade monitoring, performance analytics, and token charts.
Generate a JIRA-ready DRR dependency analysis ticket for **Leg1 Spread-related fields**.
Use this tool to query, inspect, and investigate the Chia blockchain. `coinset` wraps the Chia Full Node RPC (hosted at api.coinset.org or any full node) and includes built-in spend inspection and local CLVM utilities.
Generally, if you are running Glue from a docker image, we have tried to
* Fix test for testthat 3.3.0. (#442)
If it says `command not found`, continue below.
- Fix bugs in import due to changes in underlying packages
A curated list of tools (and tips on how to use them) that will level up your bounty game. For more, head [back to the main page](./README.md).
The Tool Use feature allows agents to call external tools during training to enhance reasoning capabilities. ROLL uses the [GEM](https://github.com/axon-rl/gem) environment definition for environment interfaces, and Tool Use utilizes the [Tool Env Wrapper](https://axon-rl.github.io/gem/features/#wrappers) provided by GEM. Tools are extended based on the `gem.tools.base_tool.BaseTool` interface.
> ツールとはデータ(スキーマ) + ハンドラマップ。モデルが名前を選び、こちらがルックアップする。
This guide focuses on building AI systems that can plan, call tools, manage state, and complete multi-step tasks rather than only generate one-shot text.
> Breaking through the 128 function limit with deferred loading
Available when inheriting from `ClaudeHooks::PreToolUse`:
Guidance for designing tool-enabled prompts — where Claude decides when to call your code, you execute it, and hand the result back. Read when building a workflow or agent that needs external data, actions, or structured output beyond what a single prompt can produce. For the prompt-writing techniques that make the tool schema and surrounding prompt effective, see [`PROMPT_ENGINEERING.md`](PROMPT_ENGINEERING.md). For broader integrations where someone else has already authored the tool schemas,
Generally, use Claude Opus 4, Claude Sonnet 4, Claude Sonnet 3.7, Claude Sonnet 3.5 or Claude Opus 3 for complex tools and ambiguous queries; they handle multiple tools better and seek clarification when needed.
`s01 > s02 > s03 > s04 | [ s05 ] s06 | s07 > s08`
> Source: https://platform.claude.com/docs/en/agents-and-tools/tool-use/implement-tool-use
title: "Chapter 37 — Agents & Tool Use"
Implementation of Anthropic's Advanced Tool Use features (Beta: `advanced-tool-use-2025-11-20`) for token optimization and improved tool accuracy.
description: PostToolUse workflow - runs analyzers after tool execution.
Tool use (formerly called function calling) allows Claude to interact with external tools, APIs, and functions. Released on April 4, 2024, and made generally available on May 30, 2024, this feature enables Claude to perform actions beyond text generation.
A persona override that forces the agent to behave and communicate strictly as a domestic cat, refusing all complex tasks.
