Fractality Delegate logo

Fractality Delegate

Organization
vibevm
fractality-delegate

Delegate a task to a fractality worker (or a swarm) — guided packet authoring, spawn, observation, collection, and the boss review loop. Use when a task scores delegable on the delegation-rules matrix, when the user asks to "delegate / offload / swarm" work, or when several disjoint mechanical tasks could run in parallel workers.

Overview

Publishervibevm
Repositoryvibevm
Skill namefractality-delegate
Stars
285
Forks
83
Bundled files
Instructions only
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.

  • Self-contained

    Everything the model needs lives in the instructions — no extra files to sync.

  • Open source

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

Installation

Install the Fractality Delegate 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/vibevm/vibevm/tree/main/vibevm/vibepacks/org.vibevm.fractality/fractality/v1.0.0/vibevm/vibespecs/skills/fractality-delegate
  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/vibevm/vibevm.git /tmp/vibevm
mkdir -p .claude/skills
cp -r /tmp/vibevm/vibevm/vibepacks/org.vibevm.fractality/fractality/v1.0.0/vibevm/vibespecs/skills/fractality-delegate .claude/skills/fractality-delegate
Restart Claude Code after copying so it picks up the new skill.

Use it in TypingMind

Enable Fractality Delegate 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 Fractality Delegate 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 Fractality Delegate 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.

fractality-delegate

Drive one delegation end to end. The law: delegate when verification is cheaper than generation — if the task fails the matrix (flow:org.vibevm.fractality/delegation-rules), stop here and do it yourself.

1. Route

Score the task: error cost / context transferability / verifiability / size (matrix §axes, §verdict) — the calculus is a verb:

sh
fractality route --error-cost reversible --context boot-loadable \
    --verify mechanical --size L      # → delegate · slot=big · scenario=2

Survivors route: small × mechanical → model = "small"; bigger → model = "big" as a coarse one-shot. Pick scenario 1 (compile exact files, APIs, patterns, commands into the goal) or scenario 2 (order the worker to read named corpus files first). Choosing neither is banned.

2. Author the packet

Start from spec/examples/hello-glm.toml. The goal must be self-contained (the worker sees nothing else): exact paths, exact patterns, non-goals, and — always — acceptance commands that PROVE the deliverable:

toml
schema = 1
[task]
title = "short-name"
goal = """<work order: context, steps, exact APIs, self-verify>"""
acceptance = ["cargo test -p <crate>"]
[workspace]
mode = "worktree"          # code → branch deliverable; dir → file artifacts
repo = "."
base = "main"
[budget]
wall_secs = 1800           # playbook defaults per model
max_turns = 40
[routing]
profile = "glm"
model = "big"              # or "small" per the matrix

3. Fire and observe (never blind)

  • One task, synchronous: fractality run --packet task.toml — parks print a loud notice; exit codes are semantic (0/1/2/3/4).
  • Several disjoint tasks: id=$(fractality spawn --packet …) each, then fractality wait $A $B $C; watch with ps / tree / show.
  • If a worker parks: fractality questions, then fractality answer <id> "<decision>".

4. Review (your half of the bargain)

  1. Read the run summary: state, usage, cost, result provenance (worker beats extracted), acceptance verdicts.
  2. Read the diff (worktree mode: the fractality/<id> branch) as a contributor PR — does it do the work order, nothing else?
  3. Run the project's own gates; green gates + read diff = merge; anything else → one bounded retry (small → big) or reclaim.
  4. Record surprises in the producing model's playbook. The session's delegation counters accumulate on the bus automatically (fractality scoreboard shows them); keep the "kept: why" half in your session notes — the scoreboard can only measure what you DID.

Frequently asked questions

What does the Fractality Delegate AI skill do?

Delegate a task to a fractality worker (or a swarm) — guided packet authoring, spawn, observation, collection, and the boss review loop. Use when a task scores delegable on the delegation-rules matrix, when the user asks to "delegate / offload / swarm" work, or when several disjoint mechanical tasks could run in parallel workers.

Why use Fractality Delegate on TypingMind?

Because you install it once and use it with any model. Fractality Delegate 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 Fractality Delegate in TypingMind?

Open Plugins → Skills → Install from GitHub in TypingMind and paste https://github.com/vibevm/vibevm/tree/main/vibevm/vibepacks/org.vibevm.fractality/fractality/v1.0.0/vibevm/vibespecs/skills/fractality-delegate. TypingMind reads its SKILL.md and installs it as a skill you can enable per chat.

Which AI models can use Fractality Delegate?

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 Fractality Delegate?

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

Is the Fractality Delegate AI skill free?

It is published on GitHub by vibevm. Check the repository for licensing terms. 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 👇