generic-assistant

Generates system prompts and configurations for general-purpose personal assistant agents.

27.6k|2.7k|Updated Aug 6, 2024
One-click install
npx skills add https://github.com/mastra-ai/mastra --skill generic-assistant
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: generic-assistant
Source: https://github.com/mastra-ai/mastra/tree/main/packages/editor/src/ee/workspace/skills/generic-assistant
Command: npx skills add https://github.com/mastra-ai/mastra --skill generic-assistant

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Building a general-purpose personal assistant agent often results in vague, chatty agents that never finish tasks. This Skill provides a fallback authoring playbook that produces decisive, scoped assistant agents with clear completion criteria when no more specific archetype (coding, research, customer-support, etc.) applies.

Core Features & Use Cases

  • Archetype fallback logic: Guides selection of this playbook only after ruling out six more specific agent archetypes, preventing misuse.
  • System prompt template: Supplies a ready-to-adapt prompt structure covering scope boundaries, decision rules, output format, communication style, and refusal handling.
  • Completion criteria enforcement: Requires every produced agent to deliver a concrete result, stay concise, and cite tool sources.
  • Use Case: A user asks for an agent that summarizes their emails each morning; the playbook produces an "Inbox Summary Helper" agent with a Gmail tool, explicit scope, and a bullet-list output format.

Quick Start

Ask the agent builder to create a personal assistant that summarizes your daily emails, and it will apply this playbook to produce a scoped, decisive assistant agent.

Frequently Asked Questions about generic-assistant

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

FAQPage Schema
How do I build a personal assistant agent that summarizes emails?

Define a narrow scope such as today's unread threads, attach only the email tool like Gmail, and require a bullet-list output with one action item per thread. The playbook's worked example shows this exact configuration.

When should I use the generic-assistant playbook instead of other archetypes?

Use it only as a fallback after checking the coding, spreadsheet, research, customer-support, content-writer, and ops-automation archetypes. It fits genuinely general helpers like inbox summarizers or note-takers.

What makes a good system prompt for a general assistant agent?

A good prompt names the user and task list, defines explicit in-scope and out-of-scope boundaries, leads with the result, and includes completion criteria. Avoid generic phrases like "You are a helpful AI assistant."

How many tools should a generic assistant agent have?

Attach only tools the user explicitly mentioned, such as Gmail for email summarization. Loading many tools just in case causes drift, and if no tools are needed, attach none.

Why does my assistant agent keep asking clarifying questions instead of answering?

The prompt likely lacks decisiveness rules and completion criteria. Instruct the agent to pick the most likely interpretation and act, delivering a result rather than a question unless the request is impossible without clarification.