Prelude logo

Prelude

Community
dmmulroy
prelude

Prelude bootstrapping for TypeScript. Use when creating or rebuilding a prelude.ts from ambient generic helpers and types.

Overview

Publisherdmmulroy
Repositoryskills
Skill nameprelude
Stars
431
Forks
14
Bundled files
1
LicenseMIT
Links
  • Markdown instructions

    A SKILL.md file the model loads on demand, so it only costs tokens when a request actually matches.

  • Works with any LLM

    AI skills are plain Markdown, not provider-specific code, so this works with GPT, Claude, Gemini, Grok, or a local model.

  • 1 bundled files

    Scripts, templates, and references the model can read while it works. Files are read-only and never executed.

  • Open source

    Published by dmmulroy on GitHub. Read the source before you install it.

Installation

Install the Prelude AI skill in TypingMind to use it with any LLM, or drop it into another agent that reads SKILL.md.

1

Install in TypingMind

TypingMind installs a skill straight from its GitHub folder — it reads SKILL.md, bundles the resource files, and stores the result locally.

  1. Open the app and go to Plugins → Skills.
  2. Choose "Install from GitHub".
  3. Paste the skill folder URL below and confirm.
    https://github.com/dmmulroy/skills/tree/main/prelude
  4. Enable the skill in any chat where you want it available.
Plugins → Skills → Add skill → From GitHub URL, then paste the folder URL and press Continue.
2

Install in another agent

Any agent that reads the Agent Skills format can use this skill — copy the folder into that agent's skills directory.

Claude Code — .claude/skills
git clone --depth 1 https://github.com/dmmulroy/skills.git /tmp/skills
mkdir -p .claude/skills
cp -r /tmp/skills/prelude .claude/skills/prelude
Restart Claude Code after copying so it picks up the new skill.

Use it in TypingMind

Enable Prelude in any TypingMind chat and the model takes it from there. Its name and description sit in the system prompt, and the moment a request matches, the model loads the full instructions itself — you never invoke it by hand, and it costs no tokens until it is actually used.

The model loads Prelude on its own as soon as a request matches it.

Works with any AI model

AI skills are plain Markdown instructions rather than provider-specific code, so Prelude is not tied to the model it was written for. Install it once in TypingMind and use it with GPT-5, Claude, Gemini, Grok, DeepSeek, Mistral, Llama, or a local model you run yourself — all on your own API keys.

  • Loaded only when it is needed

    The system prompt carries just the name and description. The instructions are fetched on the first matching request, so an idle skill costs nothing.

  • Switch models mid-chat

    Because the skill is instructions rather than code, changing model does not break it — the next model reads the same SKILL.md.

Skill instructions

This is the SKILL.md content the model loads. Read it before installing — a skill is instructions your model will follow.

Bootstrap a TypeScript Prelude

A prelude is an explicitly imported module for ambient helpers and types: ubiquitous, domain-neutral building blocks that have no more precise owner. Use the bundled prelude.ts as the foundation, then adapt it to evidence from the target repository. Apply ../coding-standards/SKILL.md throughout.

Ambient describes a helper's role, not a TypeScript global. Keep prelude exports behind ordinary imports; do not add global declarations or global augmentation.

1. Map the repository

Read repository instructions, package manifests, TypeScript configuration, source layout, and import conventions. Locate:

  • an existing prelude or equivalent shared module;
  • files named utils, helpers, common, types, result, errors, or similar;
  • generic type aliases and tiny generic functions repeated across modules;
  • established libraries for results, schemas, redaction, branding, collections, and exhaustive matching;
  • every caller of plausible ambient helpers.

Search by both filenames and concepts. Use the ubiquitous generic helper/type categories in the coding standards as seed search terms, then inspect definitions and callers rather than classifying from names alone.

Completion criterion: Every plausible ambient definition found by repository-wide filename, symbol, and duplication scans is inventoried with its owner, callers, dependencies, and current behavior.

2. Classify every candidate

A symbol belongs in the prelude when all of these hold:

  • it is domain-neutral and useful across unrelated modules;
  • no domain, application service, adapter, protocol, or focused generic module is a more precise owner;
  • centralizing it reduces duplication or gives a ubiquitous concept one canonical implementation;
  • its dependencies are minimal, stable, and already justified by the project;
  • its behavior is small enough to understand at the import site or hidden behind a precise type.

Keep a symbol with its focused owner when it encodes domain meaning, application policy, boundary translation, framework behavior, I/O, or a cohesive generic concept such as string casing or non-trivial collection operations. Prefer an established library over a local duplicate. A prelude is a curated foundation, not a barrel or miscellaneous dumping ground.

Record one decision for every candidate: use the established library, keep the current owner, move into the prelude, merge with a template symbol, or delete as an unused duplicate.

Completion criterion: Every inventoried candidate has one decision grounded in its semantics and callers; no candidate remains classified only by its filename or name.

3. Seed from the template

Read the bundled prelude.ts completely. Copy it to the project's established shared-module location as the starting point. If a prelude already exists, merge deliberately instead of overwriting it.

Choose exactly one expected-failure foundation:

  1. When the project uses Effect, use Effect's result/error facilities and remove the template's local Result fallback.
  2. When the project uses better-result, use it and remove the local fallback.
  3. When the project uses neither, ask whether to install better-result.
    • If accepted, install it and remove the local fallback.
    • If declined, enable the template's local Result types and helpers.

Retain each other template export only when repository usage, the coding standards, or the requested foundation justifies it. Preserve compatible existing behavior when merging equivalent helpers; surface semantic conflicts rather than silently choosing one implementation.

Completion criterion: The target file is founded on the template, has exactly one result strategy, and every retained template export has an explicit justification.

4. Consolidate ambient helpers and types

Move or merge the approved repository candidates into the prelude. For each moved symbol:

  • preserve behavior unless a behavior change was requested;
  • preserve or deliberately migrate its public name and type contract;
  • update every caller to import from the prelude directly;
  • retain required JSDoc, safety comments, and targeted lint suppressions;
  • remove superseded definitions and compatibility re-exports after their callers move.

Keep the resulting module side-effect free. It must not read configuration, acquire resources, register handlers, perform I/O, contain domain/application policy, or re-export unrelated modules.

Completion criterion: Every approved candidate has one canonical definition, every caller uses it, and no removed source remains as a second source of truth.

5. Verify the foundation

Run the repository's formatter, type checker, linter, and focused tests. Search again for the old symbols, duplicate definitions, stale import paths, and broad utility files that were part of the inventory. Review every prelude export for current usage or an explicit foundational reason.

Report:

  • the chosen result strategy;
  • template exports retained or removed;
  • repository helpers moved, merged, left in place, or deleted;
  • verification commands and outcomes.

Completion criterion: Repository checks pass, every inventory decision is reflected in code, duplicate ambient definitions are gone, and every prelude export is justified.

Bundled files

The model reads these on demand while the skill is loaded. They are exposed as readable files and are never executed.

Frequently asked questions

What does the Prelude AI skill do?

Prelude bootstrapping for TypeScript. Use when creating or rebuilding a prelude.ts from ambient generic helpers and types.

Why use Prelude on TypingMind?

Because you install it once and use it with any model. Prelude is plain Markdown rather than provider-specific code, so the same skill runs on GPT-5, Claude, Gemini, Grok, or a local model — and you can switch model mid-chat without it breaking. TypingMind runs on your own API keys, so you pay providers directly instead of a per-seat subscription, and your skills and chats stay in your own storage.

How do I install Prelude in TypingMind?

Open Plugins → Skills → Install from GitHub in TypingMind and paste https://github.com/dmmulroy/skills/tree/main/prelude. TypingMind reads its SKILL.md and bundles its files and installs it as a skill you can enable per chat.

Which AI models can use Prelude?

Any model you connect in TypingMind. AI skills are plain Markdown instructions rather than provider-specific code, so GPT, Claude, Gemini, Grok, and local models can all load this skill when a request matches it.

How many AI models can I use with Prelude?

As many as you like. As long as a model supports skills, you can use Prelude with it — GPT, Claude, Gemini, Grok, DeepSeek, Mistral, Llama and more — all on TypingMind with your own API keys.

Is the Prelude AI skill free?

Yes. It is published on GitHub by dmmulroy under the MIT license. You only pay your own AI provider for the tokens you use.

What are AI skills?

An AI skill is a reusable instruction bundle that teaches an AI model how to do one specific task. It follows the open Agent Skills format: a SKILL.md file with a name and description, plus any scripts, templates or reference files the model may need. The model reads the instructions only when your request matches the skill, so an installed skill costs nothing until it is used.

How are AI skills different from plugins or MCP servers?

A plugin or MCP server gives a model new tools to call — code that runs somewhere and returns a result. An AI skill gives the model knowledge and process instead: how to approach a task, which steps to follow, what good output looks like. Skills are plain Markdown, so they need no server, no API key and no runtime, and they work with any model.

View all

Set up your own AI workspace now

Get notified about new features and future giveaways by subscribing to our newsletter 👇