🤖
AI EngineeringRust

Ultra Task

by KonghaYao

Ultra Task is an AI Engineering skill for Claude Code, published by KonghaYao in peri.

163 stars30 forkson KonghaYao/periAdded 2026/09/08+3% in starsRepository updated 2026/09/07
acpclaude-codecoding-agentdeepseekglmmcpruststreaming-markdowntui
Install in seconds
Install Ultra Task
Copy Ultra Task into your Claude Code skills folder. Run the command in your terminal, or review the source on GitHub before installing.
terminal
npx degit https://github.com/KonghaYao/peri/tree/main/peri-middlewares/src/skills/builtin/skills/ultra-task ~/.claude/skills/ultra-task

Requires Node.js. Downloads this skill only — not the rest of the repository — into your Claude Code skills folder.

Without Node.js

git clone https://github.com/KonghaYao/peri.git

Clones the whole repository, then copy the skill’s own directory into your skills folder yourself.

In this catalog

Source file
peri-middlewares/src/skills/builtin/skills/ultra-task/SKILL.md in KonghaYao/peri
Installs to
~/.claude/skills/ultra-task
Collection
One of 20 skills cataloged from this repository
Category
AI Engineering3670 skills

What Ultra Task does

Ultra Task supervises non-trivial work by delegating bounded investigation, implementation, testing, and review to subagents while the main agent coordinates scope, approvals, and verification. Use it when a task needs orchestration rather than a single direct step.

Ultra Task is cataloged under AI Engineering on DirSkills. Ultra Task comes from a repository tagged acp, claude-code, coding-agent, deepseek and glm.

Documentation

README

Ultra Task

Act as the task supervisor. The Main Agent is the control plane: it owns the user's intent, task graph, authorization boundary, integration, and completion verdict. Subagents are the execution plane: delegate most bounded investigation, implementation, testing, and review work to them.

An explicit /ultra-task invocation always selects this mode. Delegation is not a quota: keep decisions that change user intent, permissions, scope, or global architecture in the Main Agent, and perform tiny coordination or integration actions directly when delegating them would cost more than doing them.

Use the direct Agent tool. Do not switch to Workflow or Ultra-ADLC unless the user separately asks for that mode. Subagents cannot launch further subagents.

This is the opening of the README. Read the full README on GitHub.

Commands Ultra Task provides

Slash commands named in this skill’s SKILL.md, listed in the order they first appear.

  • /ultra-task

Frequently asked about Ultra Task

  • What else does KonghaYao publish alongside Ultra Task?

    Ultra Task is one of 20 skills that DirSkills catalogs from KonghaYao/peri, the repository it ships in. Its siblings there include Adaptive Devflow, Advisor Consultation and Auto Converge. Each one is a separate skill with its own page in this directory, installs the same way Ultra Task does, and is maintained by KonghaYao in that same repository. The rest of the collection is listed on the KonghaYao/peri page.

  • How does Ultra Task compare to other AI Engineering skills?

    Ultra Task ranks #3396 by stars among the 3670 AI Engineering skills in this catalog. The most-starred ones next to it are Architecture Decision Records, AI-First Engineering and Agentic OS. DirSkills ranks by the star count of the repository each skill ships in, so that order reflects how popular those repositories are rather than any review of Ultra Task against them. Open each page to compare what they document and how they install.

More from KonghaYao/peri

Ultra Task is one of 20 skills cataloged on DirSkills from KonghaYao/peri.

See all 20 skills
🔧
1h ago

Adaptive Devflow

Adaptive Devflow helps choose a lightweight-to-heavy development workflow for issues, bugfixes, features, and refactors. It guides when to use coordination, review, and verification phases based on task risk and complexity.
Automation
16330
🧭
1h ago

Advisor Consultation

Advisor Consultation uses a tool-free Opus second opinion to evaluate high-risk design trade-offs, unresolved failures, and cross-boundary decisions. Use it after gathering repository evidence when the main agent needs a careful review.
AI Engineering
16330
🔁
1h ago

Auto Converge

Auto Converge refines rules, prompts, and SKILL.md files until agents can follow them reliably. It uses external references and adversarial review loops when a spec is written but not being obeyed.
AI Engineering
16330
🛠️
1h ago

Auto Issue Fixer

Auto Issue Fixer manages an issue from creation through fix, verification, and archival. Use it when someone describes a bug, asks to record a problem, fix an issue, verify a change, or archive a resolved issue.
Automation
16330
📝
1h ago

Blog Writer

Blog Writer provides a style guide for writing Peri project blog posts. Use it when drafting project intros, technical retrospectives, architecture discussions, or performance writeups in the Peri voice.
Writing
16330
🗂️
1h ago

Codebase Index

Codebase Index maps user intents like changing a threshold, tracing an event chain, or finding an entry point to the exact files and functions to open. It is used to quickly locate or rebuild per-crate code indexes for a Rust codebase.
Writing
16330