UI はローカライズ済みです。スキルの紹介と説明文書は選択した言語へ AI が自動翻訳し、コード・コマンド・パス・専門用語は原文のままです。原作者の文書が優先されます。
スキルを探す
生産性・共同作業 · SKILL.md

handling-commands

handling-commands

Patterns for handling commands, validating input, and filtering messages in XMTP agents. Use when implementing slash commands, validators, or message filters. Triggers on command handling, input validation, or type guards.

Independent GitHub snapshot · Automated classification · Original documentation governs

スキルの説明

GitHub

name: handling-commands description: Patterns for handling commands, validating input, and filtering messages in XMTP agents. Use when implementing slash commands, validators, or message filters. Triggers on command handling, input validation, or type guards. license: MIT metadata: author: xmtp version: "1.0.0"

XMTP commands and validation

Best practices for handling commands, validating input, and filtering messages.

When to apply

Reference these guidelines when:

  • Implementing slash commands
  • Validating hex strings and addresses
  • Filtering message types
  • Using type guards instead of type assertions

Rule categories by priority

PriorityCategoryImpactPrefix
1ValidatorsCRITICALvalidators-
2FiltersHIGHfilters-
3Type GuardsHIGHguards-

Quick reference

Validators (CRITICAL)

  • validators-hex - Use validHex() for hex string validation
  • validators-address - Validate Ethereum addresses

Filters (HIGH)

  • filters-message-types - Filter by message type
  • filters-sender - Filter out self-messages
  • filters-content - Check for defined content

Type Guards (HIGH)

  • guards-codec - Use usesCodec() instead of type assertions
  • guards-content-type - Use filter helpers for type safety

How to use

Read individual rule files for detailed explanations:

rules/validators-hex.md
rules/filters-message-types.md
rules/guards-codec.md
保存したスキル