aios-work-dispatch

aios-work-dispatch is a skill for Claude Code, Codex from rexleimo/aios. It costs 70 tokens per session (1,335 once invoked), scanned A, original, MIT.

A routing guide for deciding whether independent coding tasks can be sent to multiple agents at once.

In plain words
What is it for?
Use it to assess parallel-work eligibility, preview a dispatch, and require approval before starting multiple model runs.
Why use it?
It prevents parallel work when tasks share files, depend on one another, lack clear ownership, or need a single continuous process.

Skill for Claude CodeCodex

Install

Getting it into your agent

One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.

agentmods
npx agentmods add skills/rexleimo/aios/aios-work-dispatch
Any agent
npx skills add rexleimo/aios --skill aios-work-dispatch
Clone the repo
git clone --depth 1 https://github.com/rexleimo/aios

Made for: Claude Code, Codex.

Wrote this? Show the measurements

A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.

agentmods badge for aios-work-dispatch

README.md
[![agentmods](https://agentmods.dev/badge/skills/rexleimo/aios/aios-work-dispatch.svg)](https://agentmods.dev/skills/rexleimo/aios/aios-work-dispatch)
Your own site
<a href="https://agentmods.dev/skills/rexleimo/aios/aios-work-dispatch"><img src="https://agentmods.dev/badge/skills/rexleimo/aios/aios-work-dispatch.svg" alt="Measured on agentmods" height="20"></a>
Per session 70 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 1,335 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. Scan, not verified.
Origin original No closer match found in the catalogue.
Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

ModelPer sessionOnce invoked
Fable 5 $0.00070 $0.01335
Opus 5 $0.00035 $0.00668
Sonnet 5 $0.00014 $0.00267
Haiku 4.5 $0.00007 $0.00134

Measured 3d ago against content hash b6bbdee79613, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

aios-work-dispatch scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 3d ago.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Nothing flagged

None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.

skill-sources/aios-work-dispatch/SKILL.md · 148 lines

How it starts

The opening of the file, as written. The whole thing — 148 lines — stays where its author put it; the contents beside it link to each section on GitHub.

AIOS Work Dispatch

Use this Skill when deciding whether current coding task should enter aios work parallel dispatch. It explains routing; it does not replace current AIOS workflow policy or Rex Command.

Trigger eligibility

Use aios work candidate route only when all conditions hold:

  1. Current AIOS disposition is planned.
  2. Task decomposes into at least two independently executable work items.
  3. Each work item has explicit owner/path boundary and observable acceptance criteria.
  4. Work-item file ownership does not overlap.
  5. Work items do not require strict predecessor ordering.
  6. Task is not one continuous resumable objective.

If any condition is unproven, keep execution serial and do not launch aios work.

Do not use

  • Small local change or typo: use current single-agent route.
  • Coupled edits or strict ordered phases: use current Provider or aios orchestrate.
  • One long resumable objective: use aios harness.
  • Status, history, or observability only: use aios team.
  • Review-only request: do not dispatch implementers.
  • Unclear ownership, missing acceptance criteria, or unknown client readiness: stop and clarify or remain serial.

Required execution sequence

0. Plan the decomposition first (user need -> work items)

aios work decomposes from a structured plan when one is active, otherwise from the --context string. Plan first so the engine can use real dependencies and ownership instead of guessing:

aios plan start --title "<objective>" --task "<user need>"
aios plan task <id> --target <path> --allow-write <glob>   # declare ownership
aios plan task <id> --context <ref>                         # per-task context

Independent tasks get empty dependsOn; coupled tasks declare their prerequisite ids. Tasks that share file ownership are NOT independent — merge gate blocks them. After the plan has at least two eligible (pending) tasks, dispatch reads it:

aios work --task "<objective>" --dry-run --json

Read the full file on GitHub · 148 lines

Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

  1. 3d ago First seen · 148 lines · 70 tokens per session scan A b6bbdee79613

Subscribe to this mod's changes

aios-work-dispatch is a skill published in the GitHub repository rexleimo/aios (52 stars, last pushed 4d ago), licensed MIT. It adds 70 tokens to every session and 1,335 once invoked, about $0.0003 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.