Agent skill

Multitask

by KonghaYao in KonghaYao/peri

You are the coordinator: every workstream leaves your hands and belongs to exactly one owner launched through Agent, and coordination is the only work that stays with you.

Apache-2.0Auto-check passed

Install Multitask

skills CLI
$ npx skills add KonghaYao/peri --skill multitask -a claude-code

Project install by default; add -g for ~/.claude/skills/.

GitHub CLI
$ gh skill install KonghaYao/peri multitask --agent claude-code

Project scope by default; add --scope user for a personal install. Needs GitHub CLI 2.90.0 or later (public preview).

Manual copy
$ git clone --depth 1 https://github.com/KonghaYao/peri.git skills-src && mkdir -p .claude/skills && cp -r skills-src/peri-middlewares/src/skills/builtin/skills/multitask .claude/skills/multitask && rm -rf skills-src

Use ~/.claude/skills/ instead of .claude/skills for a personal install. The folder must contain SKILL.md.

Claude Code skills documentation · loads skills from .claude/skills/

Facts

Skill name
multitask
GitHub stars
223
Token cost
~1.4k tokens
SKILL.md length
868 words
Files
1
Skills in repo
19
Repo updated
First seen
Licence
Apache-2.0

At a glance

You are the coordinator: every workstream leaves your hands and belongs to exactly one owner launched through Agent, and coordination is the only work that stays with you.

  • SKILL.md covers What this mode changes, One owner, Background only and Todo as the record, plus 3 more sections
  • Instructions only: no scripts, shell commands, URLs or credentials in SKILL.md

What it does

Multitask is an agent skill from KonghaYao/peri. You are the coordinator: every workstream leaves your hands and belongs to exactly one owner launched through Agent, and coordination is the only work that stays with you. In force for the rest of the session once loaded.

Its SKILL.md is about 1.4k tokens, which your agent loads only when the skill is triggered. It is a single SKILL.md file with no bundled scripts.

The repository describes itself as: Lightweight Rust Agent only use 50MB RAM, but Claude Code Plugin compatible, Dynamic Workflow, Goal, Artifacts, Free Web Search, full feature and better support! The licence is Apache-2.0.

Example prompts

  • “/multitask”

What it can do on your machine

Read from SKILL.md and the folder at commit d7ee444. It shows what the files ask for, not the result of running them.

  • Tool permissions

    Pre-approves nothing: there is no allowed-tools line, so your agent's usual permission prompts apply.

    From allowed-tools in the SKILL.md frontmatter.

  • Runs code

    No scripts in the folder and no shell commands in SKILL.md.

    From the folder's file list and the shell code blocks in SKILL.md.

  • Network

    No URLs in SKILL.md.

    From URLs in SKILL.md, links to its own repository left out.

  • Credentials

    Names no API keys, tokens, secrets or passwords.

    From names ending in _API_KEY, _TOKEN, _SECRET, _KEY or _PASSWORD in SKILL.md.

Context cost

Multitask loads about 1.4k tokens when it runs. Until then it costs about 58 tokens; SKILL.md has 868 words of instructions outside code blocks.

Always · name and description, kept in context so the agent knows when to use it
~58
When it runs · the whole SKILL.md, loaded when a task matches
~1.4k

Estimates: characters ÷ 4, the usual rule of thumb; real counts depend on the model's tokenizer. Scripts and assets cost tokens only if the agent reads them.

Safety

Auto-check passed

The automated check found no risky patterns in SKILL.md.

Automated static check — not a guarantee. Review scripts before installing. It scans the text of SKILL.md for risky patterns (piping downloads into a shell, reading credential files, hidden Unicode, destructive commands); files beside SKILL.md are not scanned.

SKILL.md

The full file from KonghaYao/peri at commit d7ee444, republished under its Apache-2.0 licence (© KonghaYao). 868 words, ~1,445 tokens.

Download SKILL.mdSave it as .claude/skills/multitask/SKILL.md (or your agent's skills folder).
name
multitask
description
You are the coordinator: every workstream leaves your hands and belongs to exactly one owner launched through `Agent`, and coordination is the only work that stays with you. In force for the rest of the session once loaded.

Multitask Mode

You are the coordinator. Every workstream leaves your hands and belongs to exactly one owner launched through Agent. The mode holds from the moment you have it, for the rest of the session — there is no opt-in, no task it exempts, and no small change it lets you keep.

What this mode changes

The host default lets you work in the foreground whenever the change fits, and puts direct work at two or three files. Under this mode none of that zone survives: every deliverable goes to an owner.

Coordination is the only work left to you — reading owner reports, synthesizing, asking the user, deciding the next step, launching and resuming owners, and keeping the record below. Everything else is a deliverable:

  • Any read, search, or edit of project material. One file is already enough.
  • Recon that enumerates a directory, greps the tree for references, or reads build, deploy, or CI config to pin down a vague brief. Recon is a deliverable like any other; hand it off, and the work it sizes stays with that owner.
  • Any change crossing a crate boundary.
  • Two or more independent workstreams the user hands over — one owner each.

An owner's returned report is coordination; the files behind it are not. Where the line is unclear, hand off rather than take a look first.

If Agent is unavailable, or the user prohibits delegation, do the permitted work directly and say so. Never refuse work the user has authorized.

One owner

A deliverable you hand off — investigate → implement → verify — has exactly one owner, and that owner is never you. Absorbed:

  • Do not redo handed-off work yourself or in a second owner; follow up only where results expose a gap.
  • Verification belongs to the deliverable. Read the evidence your owner reports instead of re-running its verification. When a real gap appears, hand the gap back to that owner (resume_thread_id) — a gap that falls outside its original file scope is still the owner's to close, not yours.
  • Do not split one deliverable into roles that wait on each other.
  • Siblings only for genuinely independent workstreams, with non-overlapping write scopes and settled interfaces. You hold no files, and an owner's files are not yours to edit; wait, or hand ownership over explicitly.
  • Delegation is one level deep, so you alone add siblings. Independent siblings run concurrently — launch them together and wait for their completion notifications instead of serializing the work or taking it back.

Before handing off: does this work already have an owner, and does the file scope you are about to assign overlap a sibling's? Then give the brief a file scope, constraints, acceptance criteria, and the expected return. A launch acknowledgment is not delivery.

Background only

Every hand-off launches with run_in_background: true; you never block on a synchronous Agent call. Blocking trades away the very freedom the hand-off exists to preserve, so no hand-off is the exception.

Show full SKILL.md (384 more words)Show less

Todo as the record

Background hand-offs give you nothing to await, so the todo tool (TodoWrite) is the record of what is in flight: one item per handed-off deliverable, opened at launch and closed only on evidence.

  • Write the deliverable and the owner's child_thread_id into the item. That is what makes a later follow-up a resume rather than a re-briefing.
  • Items are hand-offs and coordination steps, never a mirror of an owner's plan — the owner keeps its own list, and a duplicate is not a record.
  • An item closes on the owner's returned report or a completion notification. Nothing weaker — not the launch, not a queued prompt, not an assumed success.
  • Blocked, interrupted, and cancelled work keeps its item, carrying the blocker. Deleting it loses both the thread and the reason; clear the blocker with resume_thread_id.
  • What the user asks next is answered from this list: what is still running, what came back, what is waiting on them.

Model choice

  • Lookup and mechanical sweeps → haiku. Implementation with verification → sonnet, or omit model and take the definition's tier. Cross-crate contracts and architecture trade-offs → opus. The user asks for the parent's model → inherit.
  • Do not cut implementation work to haiku to save cost — that trades the price of delegation for a weaker executor.
  • resume_thread_id and fork reuse the original environment and ignore model, so never drive a model change through either. Do not interrupt a running owner to switch tiers.

Follow-up

  • Keep the child_thread_id and follow up on the same work through Agent(resume_thread_id: ...) rather than creating a replacement owner; read the returned action to tell sending from resuming.
  • Interrupted with work left and no cancellation: resume that thread. If it is still active but has no live receiver in this session, report the blocker rather than silently recreating it.
  • A sent prompt is queued rather than interrupting an in-flight call. Until it appears in the transcript, do not claim the changed requirement is implemented.
  • On completion, do only gap follow-up and one synthesis for the user: changes, verification evidence, blockers, unverified items.

Cancellation

Background owners have their own lifecycle, so cancelling your turn does not stop them. Use the host's cancellation interface and confirm the result. Cancelled work is not resumed automatically. Evidence of state is the actual tool response or runtime notification, never an assumption.

© KonghaYao, Apache-2.0. Rendered from Markdown: HTML in the file is shown as text, images as links, and headings moved down two levels. Raw file

Files

Just SKILL.md in peri-middlewares/src/skills/builtin/skills/multitask of KonghaYao/peri.

Open the folder on GitHubat commit d7ee444

Compare with similar skills

Multitask next to the 5 skills that share the most tags, products or categories with it. Stars are the repository's; “used in” counts other GitHub owners with a copy.

Multitask compared with similar skills
SkillStarsUsed inTokensAuto-checkLicenceRepo updated
Multitask this skillKonghaYao/peri223—~1.4kAutomated safety check: PassApache-2.0
Leave Managementsickn33/agentic-awesome-skills47k1 repos~3.7kAutomated safety check: PassMIT
Agent Consensus Coordinatorruvnet/ruflo74k3 repos~3.2kAutomated safety check: PassMIT
Agent Mesh Coordinatorruvnet/ruflo74k3 repos~3.2kAutomated safety check: PassMIT
Agent Queen Coordinatorruvnet/ruflo74k3 repos~1.3kAutomated safety check: PassMIT
Agent Adaptive Coordinatorruvnet/ruflo74k2 repos~4kAutomated safety check: PassMIT

Similar skills

  • Leave Management

    sickn33/agentic-awesome-skills

    Leave register: request, leave type, employee and department, manager and approver, start and end dates, days requested, leave balances, handover notes and status.

    47k GitHub starsUsed in 1 repo~3.7k tokens
    Auto-check passed
  • Agent skill for consensus-coordinator - invoke with $agent-consensus-coordinator

    74k GitHub starsUsed in 3 repos~3.2k tokens
    Agent WorkflowsAuto-check passed
  • Agent skill for mesh-coordinator - invoke with $agent-mesh-coordinator

    74k GitHub starsUsed in 3 repos~3.2k tokens
    Agent WorkflowsAuto-check passed
  • Agent skill for queen-coordinator - invoke with $agent-queen-coordinator

    74k GitHub starsUsed in 3 repos~1.3k tokens
    Auto-check passed
  • Agent skill for adaptive-coordinator - invoke with $agent-adaptive-coordinator

    74k GitHub starsUsed in 2 repos~4k tokens
    Data & AnalyticsAuto-check passed
  • Agent skill for hierarchical-coordinator - invoke with $agent-hierarchical-coordinator

    74k GitHub starsUsed in 2 repos~2.8k tokens
    Agent WorkflowsAuto-check passed

More from KonghaYao/peri

All 19 skills in this repo
  • Queries Langfuse traces, prompts, datasets and sessions, and analyzes local LLM gateway logs for requests, context growth, token use and cache hits.

    223 GitHub stars~4.3k tokensUpdated today
    Auto-check: notes
  • Audits recent agent conversation history and turns repeated failures and successes into testable harness improvement proposals that later audits can check.

    223 GitHub stars~3.5k tokensUpdated today
    Auto-check passed
  • Runs commands, reads and edits files, and copies data on remote machines through a single-file Node script that wraps the system ssh and scp, in Chinese.

    223 GitHub stars~924 tokensUpdated today
    Auto-check: warnings
  • Advisor Consultation

    KonghaYao/peri

    Sends a compact, redacted decision packet to a tool-free Opus advisor subagent when a task has high-risk trade-offs or stalled investigations, then weighs the answer.

    223 GitHub stars~1.3k tokensUpdated today
    Auto-check passed
  • Scheduled Tasks Cron

    KonghaYao/peri

    Registers, lists and removes recurring agent tasks with five-field cron expressions, and sets safety rules so a schedule is created only when the user clearly asks.

    223 GitHub stars~683 tokensUpdated today
    Auto-check passed
  • Verifies and repairs a feature by using the real Peri terminal UI as a user would, looping verify, decide, fix and review until a fresh round shows no blockers.

    223 GitHub stars~2.1k tokensUpdated today
    Auto-check: notes

Questions about Multitask

What does Multitask do?

You are the coordinator: every workstream leaves your hands and belongs to exactly one owner launched through Agent, and coordination is the only work that stays with you. Multitask is an agent skill from KonghaYao/peri. You are the coordinator: every workstream leaves your hands and belongs to exactly one owner launched through Agent, and coordination is the only work that stays with you.

How do I install Multitask in Claude Code?

Run `npx skills add KonghaYao/peri --skill multitask -a claude-code`. Or copy the skill folder (peri-middlewares/src/skills/builtin/skills/multitask in KonghaYao/peri) into .claude/skills/multitask in your project. Claude Code loads it when a task matches its description.

How do I install Multitask in Codex?

Run `npx skills add KonghaYao/peri --skill multitask -a codex`. Or copy the skill folder (peri-middlewares/src/skills/builtin/skills/multitask in KonghaYao/peri) into .agents/skills/multitask in your project. Codex loads it when a task matches its description.

Can I use Multitask in Cursor, Gemini CLI or GitHub Copilot?

Cursor, Gemini CLI, GitHub Copilot and OpenCode also load SKILL.md folders. With the skills CLI, run `npx skills add KonghaYao/peri --skill multitask -a cursor` (or -a gemini-cli, github-copilot or opencode for the others). To copy it by hand, put the folder in .cursor/skills/multitask, .gemini/skills/multitask, .github/skills/multitask and .opencode/skills/multitask in your project.

What does Multitask need to run?

SKILL.md names no scripts, command-line tools or credentials: Multitask is instructions for the agent only.

Does Multitask access the network?

SKILL.md contains no URLs. Any network use would come from the scripts or tools the agent runs. This is read from the text; nothing was executed.

Is Multitask safe to install?

Our automated static check of SKILL.md found no risky patterns, such as piping downloads into a shell, reading credential files or hidden Unicode. It is not a guarantee. Review the folder before installing.

What licence does Multitask use?

Multitask is published under the Apache-2.0 licence (the repository's licence). It allows redistribution, so the full SKILL.md is shown on this page.

How many tokens does Multitask use?

About 1.4k tokens (SKILL.md is roughly 5.8k characters). Agents keep only the skill's name and description in context until a task matches; then they load SKILL.md in full.

What are the alternatives to Multitask?

Skills that share tags, products or a category with Multitask: Leave Management (sickn33/agentic-awesome-skills, 47k stars), Agent Consensus Coordinator (ruvnet/ruflo, 74k stars), Agent Mesh Coordinator (ruvnet/ruflo, 74k stars) and Agent Queen Coordinator (ruvnet/ruflo, 74k stars). The comparison table on this page puts their stars, adoption, token cost, safety result and licence side by side.

Who maintains Multitask?

KonghaYao (a GitHub user) maintains it in KonghaYao/peri, which has 223 GitHub stars. The repository holds 19 skills in this directory. The repository was last updated on October 7, 2026.

Source: KonghaYao/peri on GitHub. Facts on this page come from the repository at the commit we read; the author's words are quoted as theirs.