A command for turning a problem definition into an incremental development task using the AB Method. It interviews you, checks project documents, breaks the work into missions, and runs each mission through TDD, meaning tests are written before implementation.

Use it to define a new task, create a progress tracker, plan small missions, implement them with test-first development, and review the final code changes.

It reduces ambiguity before coding and keeps progress, project rules, tests, and review in one workflow. It also surfaces conflicts and findings that need your attention.

create-task — Command by ayoubben18

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/ayoubben18/ab-method/create-task

Clone the repo

git clone --depth 1 https://github.com/ayoubben18/ab-method

Made for: Claude Code.

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 576 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.

Model	Per session	Once invoked
Fable 5	$0.00000	$0.00576
Opus 5	$0.00000	$0.00288
Sonnet 5	$0.00000	$0.00115
Haiku 4.5	$0.00000	$0.00058

Measured 2d ago against content hash 510e90bd05eb, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

create-task 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 2d 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.

.claude/commands/create-task.md · 42 lines

How it starts

The opening of the file, as written. The whole thing — 42 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Create Task

Description

Create a new task from a problem definition using the AB Method's incremental tasking workflow system.

Usage

/create-task

Behavior

Loads and executes the create-task workflow from .ab-method/core/create-task.md

This workflow will:

Always invoke grill-with-docs to interview the user (no skip — even when the request looks clear)
Read UBIQ + CONTEXT + tech-stack + patterns + ADRs to ground the task in canonical terms
Write a slim progress-tracker.md with all missions defined as one-line entries
Run critique-plan on the drafted missions — a read-only domain critic that pushes back only on genuine conflicts with the domain model (advisory; silent when the plan is sound)
Run every mission through the tdd skill (red-green-refactor) — no separate mission docs are created
Run review-implementation after the last mission — three critics (cleaner-architecture, slop-defender, reusability-inspector) on the task diff; interactive here, so findings are presented for you to apply

Workflow Details

Always grill — grill-with-docs runs on every invocation
Always TDD — every mission goes through the tdd skill; the test is the spec
No mission docs — missions live as one-line entries in progress-tracker.md; tight technical summaries are appended on completion
Subagents only when warranted — direct implementation by default; pick agents by need (backend / UI / testing / quality / research), not by mission type
Optional parallel groups — independent missions can be tagged [pp-1], [pp-2], ... and run concurrently in subagents; untagged missions stay sequential and act as barriers. Strictly opt-in: the workflow always asks before tagging anything
Unresolved questions (rare) — a question you genuinely can't answer yet gets parked in docs/tasks/<task>/unresolved-questions.md with an agreed placeholder, instead of stalling the task or letting the agent guess. Missions that build on one are marked ⚠️ UQ-n and ship the placeholder behind a single TODO(UQ-n) seam; answer it later and /extend-task swaps it in

Read the full file on GitHub · 42 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.

2d ago First seen · 42 lines · 0 tokens per session scan A 510e90bd05eb

Subscribe to this mod's changes

create-task is a command published in the GitHub repository ayoubben18/ab-method (187 stars, last pushed 1mo ago), licensed MIT. It costs nothing until one of its globs matches a file; then it loads 576 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 2026-08-30.

Code review Planning Testing

create-task

Getting it into your agent

What it costs to keep this loaded

Grade A, and why

Nothing flagged

How it starts

Create Task

Description

Usage

Behavior

Workflow Details

What this file has done since we first saw it

Other commands, from other repositories

checklist

clarify

specify

analyze

converge

implement