prompt-builder

Construct modular, context-aware system prompts for AI agents in Markdown.

6|2|Updated Jan 29, 2026
One-click install
npx skills add https://github.com/filipexyz/ravi --skill prompt-builder-filipexyz
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: prompt-builder
Source: https://github.com/filipexyz/ravi/tree/main/src/plugins/internal/ravi-dev/skills/prompt-builder
Command: npx skills add https://github.com/filipexyz/ravi --skill prompt-builder-filipexyz

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This skill solves the challenge of maintaining consistent, context-aware system prompts across different communication channels and agent sessions, preventing fragmented or stateless AI behavior.

Core Features & Use Cases

  • Context Injection: Automatically merges identity, workspace instructions, and channel-specific formatting into a unified Markdown prompt.
  • Cross-Platform Adaptation: Dynamically adjusts output formatting based on the destination platform, such as WhatsApp, Matrix, or TUI.
  • Use Case: When an agent needs to respond in a group chat, this skill ensures the agent understands the group context, follows platform-specific formatting rules, and uses silent tokens when no response is required.

Quick Start

Ask the prompt builder to generate the current system prompt configuration for the main agent session.

Frequently Asked Questions about prompt-builder

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

FAQPage Schema
How do I standardize system prompts for AI agents across different platforms?

Standardize AI agent prompts by aggregating identity, workspace rules, and channel-specific metadata into a unified Markdown prompt, ensuring consistent behavior across diverse communication platforms and session states.

How do I build context-aware prompts for AI agents running in group chats?

Build context-aware prompts by automatically merging agent identity and workspace instructions with channel-specific formatting rules, enabling the agent to understand group context and use silent tokens when no response is required.

Why does my AI agent behave inconsistently across different communication channels?

AI agents behave inconsistently when system prompts are stateless or fragmented; using a modular prompt builder aggregates identity and channel-specific metadata to maintain consistent agent behavior across platforms like WhatsApp and Matrix.

Does this prompt-engineering tool support cross-platform adaptation for WhatsApp and Matrix?

Yes, this prompt-engineering tool supports cross-platform adaptation by dynamically adjusting output formatting based on the destination platform, including WhatsApp, Matrix, and TUI environments.

Can I generate a system prompt configuration for a runtime environment without manual Markdown formatting?

Yes, you can generate a system prompt configuration without manual formatting by using a typed internal contract that automatically renders final instructions as clean, human-readable Markdown for the runtime environment.

What is the best way to maintain context management for AI agents across multiple session states?

The best way to maintain context management is to operate within a dedicated runtime that constructs modular system prompts, aggregating identity and workspace rules to prevent stateless AI behavior across active session states.