aoi

aoi is a skill for Claude Code from Ryan529616/aoi-orgware. It costs 81 tokens per session (1,790 once invoked), scanned A, original, MIT.

A skill for working in repositories governed by AOI, an infrastructure that records agent task ownership, permissions, evidence, and closure. It explains how to bind a session to a task and how to handle governed work.

In plain words
What is it for?
Use it when AOI is active, when creating, claiming, verifying, or closing a task, or before sending work to a governed sub-agent.
Why use it?
It reduces collisions between parallel agents and prevents tasks from being closed without evidence or without recording who owns them.

Skill for Claude Code

Written for Claude Code: SessionStart hook event. Also seen: mentions subagents.

Good fit Use it when AOI is active, when creating, claiming, verifying, or closing a task, or before sending work to a governed sub-agent.

Compare 6 skills from other repositories ↓
Install with agentmods
npx agentmods add skills/ryan529616/aoi-orgware/claude
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.

Any agent
npx skills add Ryan529616/aoi-orgware --skill claude
Clone the repo
git clone --depth 1 https://github.com/Ryan529616/aoi-orgware

Made for: Claude Code.

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 aoi

README.md
[![agentmods](https://agentmods.dev/badge/skills/ryan529616/aoi-orgware/claude.svg)](https://agentmods.dev/skills/ryan529616/aoi-orgware/claude)
Your own site
<a href="https://agentmods.dev/skills/ryan529616/aoi-orgware/claude"><img src="https://agentmods.dev/badge/skills/ryan529616/aoi-orgware/claude.svg" alt="Measured on agentmods" height="20"></a>
Per session 81 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 1,790 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. A grade says what 26 rules found in the file — not that it is safe.
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 5.1 $0.00081 $0.01790
Opus 5 $0.00041 $0.00895
Sonnet 5 $0.00016 $0.00358
Haiku 4.5 $0.00008 $0.00179

Measured 7d ago against content hash 578af1255b15, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-07, from the pricing page.

Security

Grade A, and why

aoi 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 7d 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.

src/aoi_orgware/resources/claude/SKILL.md · 156 lines

How it starts

The opening of the file, as written. The whole thing — 156 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Operating under AOI

This project is governed by AOI (Agent Organization Infrastructure). AOI is a cooperative, fail-open procedural guardrail — not a sandbox. It keeps parallel agent work from overwriting ownership, losing decisions, or being closed without evidence. Work with it: the fastest path is to follow the loop below, not to route around the hooks.

The hooks will tell you the current state. Read what they say each turn.

First: bind this session to a task

On session start the hook prints one of:

  • "AOI is active … No unambiguous task mapping exists for this session." — the repo is governed but this chat is not yet bound to a task. Before any material edit, either resume an existing task or create one, then bind:

    aoi status                      # see existing tasks
    aoi resume --task <task-id>     # if continuing one
    # …or create a new one (see the loop below), then:
    aoi bind-session --task <task-id> --session-id <this-session-id>
    
  • "This session is bound to task …" — run aoi resume --task <id> and reconstruct from the task checkpoint and current files, not from memory.

Read-only answers (explaining code, summarizing) do not require a task. Only bind before material mutation or an external action.

The governed task loop

# 1. Create a task and make its plan explicit.
aoi init-task --task-id <id> --title "<title>" --objective "<why>" \
  --owner root --completion-boundary "<what 'done' concretely means>"
$EDITOR .aoi/tasks/<id>/plan.md
aoi approve-plan --task <id> --note "Scope and verification are explicit"

# 2. Claim the EXACT write scope before you mutate anything.
aoi claim --task <id> --token <id>-claim --owner root --kind implementation \
  --lock repo:file:path/to/file --intent "<what/why>" \
  --validation "<how it will be checked>" --expires-at <tz-aware-timestamp>

# 3. Do the work, then record real evidence (see "Evidence" below).

# 4. Account for delivery, release the claim, checkpoint, and close.
aoi set-delivery --task <id> --mode local-only --detail "<where the change lives>"
aoi release-claim --token <id>-claim --status done --reason "<done>"
aoi checkpoint --task <id> --next-action "Close the task"
aoi close-task --task <id> --summary "<result>"

Read the full file on GitHub · 156 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. 7d ago First seen · 156 lines · 81 tokens per session scan A 578af1255b15

Subscribe to this mod's changes

aoi is a skill published in the GitHub repository Ryan529616/aoi-orgware (2 stars, last pushed 14d ago), licensed MIT. It adds 81 tokens to every session and 1,790 once invoked, about $0.0004 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 2026-08-31.