agentleFS
Sign inStart free
albkhachikyanalbkhachikyan/claude-code-task-splitter/llms.txt

claude-code-task-splitter

llms.txt2 starsChanged 9 days agoMITView on GitHub

Task Splitter is a Claude Code skill and plugin that lowers the model cost of building features. It splits a task into atomic subtasks, scores each one's difficulty on five dimensions, and routes it to the cheapest capable tier: inline shell (T0), Haiku low (T1), Sonnet medium (T2), Sonnet high (T3) or Opus high (T4). It batches subtasks with the same model and effort into one subagent, runs independent batches in parallel waves, verifies with the project's own tests and…

  • In this repositoryOne other file it gives its agents.

The file

Create a free account to mount
Files
llms.txt
# Task Splitter

> Task Splitter is a Claude Code skill and plugin that lowers the model cost of building features. It splits a task into atomic subtasks, scores each one's difficulty on five dimensions, and routes it to the cheapest capable tier: inline shell (T0), Haiku low (T1), Sonnet medium (T2), Sonnet high (T3) or Opus high (T4). It batches subtasks with the same model and effort into one subagent, runs independent batches in parallel waves, verifies with the project's own tests and linters, and escalates only failing subtasks by one tier.

Install: `/plugin install task-splitter --marketplace albkhachikyan/claude-code-task-splitter`. Invoke: `/task-splitter:task-splitter <task>` (plugin) or `/task-splitter <task>` (manual install). License: MIT. Author: Albert Khachikyan.

## Docs

- [README](https://github.com/albkhachikyan/claude-code-task-splitter/blob/main/README.md): overview, install, usage, tiers, comparison, FAQ
- [SKILL.md](https://github.com/albkhachikyan/claude-code-task-splitter/blob/main/SKILL.md): the full workflow and hard rules the skill follows
- [Routing rubric](https://github.com/albkhachikyan/claude-code-task-splitter/blob/main/references/routing-rubric.md): scoring dimensions, override rules, worked examples, calibration
- [Batch brief template](https://github.com/albkhachikyan/claude-code-task-splitter/blob/main/references/batch-brief-template.md): prompt skeleton and reply format for batch subagents
- [Split ledger template](https://github.com/albkhachikyan/claude-code-task-splitter/blob/main/templates/split-ledger.md): format of the resumable task ledger

## Optional

- [Next.js Frontend Foundation](https://github.com/albkhachikyan/nextjs-frontend-foundation): companion skill that scaffolds strict Next.js 16 frontends

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

Check your setup, free