haiku

haiku is a command for coding agents from ruizrica/toolkit. It costs 0 tokens per session (750 once invoked), scanned A, original, MIT.

A command that assigns a task to a coordinated team of 11 AI agents: one manager, agents that gather context, agents that implement the work, and agents that validate it.

In plain words
What is it for?
Use it for research, analysis, or implementation tasks that benefit from parallel exploration, coding, and validation.
Why use it?
It divides research, coding, and checking into separate roles so larger tasks receive context gathering and cross-checking as well as implementation.

Command

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 commands/ruizrica/toolkit/haiku
Clone the repo
git clone --depth 1 https://github.com/ruizrica/toolkit

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 haiku

README.md
[![agentmods](https://agentmods.dev/badge/commands/ruizrica/toolkit/haiku.svg)](https://agentmods.dev/commands/ruizrica/toolkit/haiku)
Your own site
<a href="https://agentmods.dev/commands/ruizrica/toolkit/haiku"><img src="https://agentmods.dev/badge/commands/ruizrica/toolkit/haiku.svg" alt="Measured on agentmods" height="20"></a>
Per session 0 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 750 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 $0.00000 $0.00750
Opus 5 $0.00000 $0.00375
Sonnet 5 $0.00000 $0.00150
Haiku 4.5 $0.00000 $0.00075

Measured 4d ago against content hash bbf0db6d1178, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

haiku 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 4d 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.

docs/commands/haiku.md · 109 lines

How it starts

The opening of the file, as written. The whole thing — 109 lines — stays where its author put it; the contents beside it link to each section on GitHub.

/haiku

Spawn a team of 10 Haiku agents managed by Opus for any task. This command provides structured multi-agent orchestration with distinct phases for context gathering, implementation, and validation.

Usage

/haiku [task description - research, analyze, implement, etc.]

Arguments

Argument Required Description
task description Yes The task for the Haiku team (research, analyze, implement, etc.)

Team Structure

The command orchestrates 11 agents total:

Role Count Purpose
Opus Orchestrator 1 Parse task, assign work, synthesize results
Context Haikus 4 Gather context before implementation
Implementation Haikus 4 Execute the main work in parallel
Validator Haikus 2 Verify and cross-check results

How It Works

Phase 1: Context Gathering

Four Haiku agents launch in parallel to gather context:

  1. Code Explorer - Find relevant files, patterns, similar implementations
  2. Architecture Scout - Map dependencies, entry points, module structure
  3. History Investigator - Check git history, recent changes, ownership
  4. Test Analyst - Find test patterns, coverage gaps, quality requirements

Phase 2: Task Decomposition

Opus analyzes all context findings and:

  • Classifies the task type (research, implementation, analysis)
  • Creates 4 specific subtasks informed by the gathered context
  • Ensures subtasks are independent and parallelizable

Phase 3: Implementation

Four Haiku agents launch in parallel with roles based on task type:

For Research Tasks:

  • Deep Dive Reader, Documentation Analyzer, Pattern Finder, Edge Case Hunter

For Implementation Tasks:

  • Core Implementer, Test Writer, API Handler, Integration Specialist

For Analysis Tasks:

  • Code Quality Reviewer, Performance Investigator, Security Auditor, Test Coverage Analyzer

Read the full file on GitHub · 109 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. 4d ago First seen · 109 lines · 0 tokens per session scan A bbf0db6d1178

Subscribe to this mod's changes

haiku is a command published in the GitHub repository ruizrica/toolkit (5 stars, last pushed 20d ago), licensed MIT. It costs nothing until one of its globs matches a file; then it loads 750 tokens. 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.