skills.rest
mpsuesser
mpsuesserCOMMUNITY
Explore all skills by this author →
GitHub

effect-ai-prompt

Constructs TypeDoc-style AI prompts for multi-turn conversations with roles, tools and attachments using Astro's API.

22|1|Updated Apr 14, 2026
Search Tags:#serialization#Effect-TS#prompt construction#language models#type-safe api#tool call history#multimodal messages
One-click install
npx skills add https://github.com/mpsuesser/pi-effect-harness --skill effect-ai-prompt-mpsuesser
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: effect-ai-prompt
Source: https://github.com/mpsuesser/pi-effect-harness/tree/main/harnesses/effect/skills/effect-ai-prompt
Command: npx skills add https://github.com/mpsuesser/pi-effect-harness --skill effect-ai-prompt-mpsuesser

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Build correct, type-safe prompt messages for Effect AI without hand-assembling malformed conversation structures, especially when mixing system/user/assistant roles and tool call/result parts.

Core Features & Use Cases

  • Type-safe Prompt API mastery: Create and compose prompts using Effect’s Prompt constructors, including full conversation history management.
  • Structured message parts: Build text, reasoning, file/attachments, tool-call, and tool-result parts with the right shapes.
  • History transformation: Convert Response parts back into prompt history to keep multi-turn interactions consistent.

Quick Start

Use the effect-ai-prompt skill to assemble a multi-turn conversation prompt that includes a system instruction, a user question, and tool call/result parts.

Frequently Asked Questions about effect-ai-prompt

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

FAQPage Schema
How do I construct type-safe LLM prompts with Effect-TS?▼

To construct type-safe LLM prompts with Effect-TS, use the Effect Prompt constructors to assemble role-based, multi-part messages. This ensures structured conversation building without hand-assembling malformed message arrays.

How do I convert Response parts into prompt history for multi-turn chat?▼

Convert Response parts into prompt history by using Prompt.fromResponseParts. This transforms previous assistant responses and tool results back into valid prompt messages, keeping multi-turn interactions consistent.

How do I integrate tool calls and tool results into language model conversations?▼

Integrate tool calls and results by using structured message parts within the Effect Prompt API. You can append tool-call and tool-result shapes alongside standard text and file attachments for multimodal messages.

Why do my hand-assembled Effect AI prompts fail when mixing system and user roles?▼

Hand-assembled Effect AI prompts fail because mixing system, user, and assistant roles requires specific type-safe shapes. Using the correct Prompt constructors prevents malformed conversation structures and tool call serialization errors.

Does the Effect Prompt API support multimodal messages with file attachments?▼

Yes, the Effect Prompt API supports multimodal messages. You can include optional file attachments within structured message parts when building language model conversations alongside text and reasoning components.

Related Skills

Explore More Software Engineering Skills→
desibarradesibarra

agent-builder-pydantic-ai

Build type-safe AI agents with Pydantic AI and tool calling.

Community
Intermediate
kcns008kcns008
5

openai-whisper-api

Transcribe audio files into text via the OpenAI Whisper API.

Community
Intermediate
andrueandersoncsandrueandersoncs
8

Streams

Create and transform Effect Streams for asynchronous data processing in TypeScript.

Community
Intermediate
PaulRBergPaulRBerg
69

refine-prompt

Optimize LLM prompts and append the final prompt to .ai/PROMPT.md.

Community
Basic