create

A command that creates a custom, iterative development workflow for one feature in a codebase. It analyzes the project, gathers the user's desired capabilities, writes a capability specification, and generates instructions for building them.

In plain words
What is it for?
Use it to create a feature specification and an auto-develop command, optionally add code instrumentation, and generate a workflow that implements new functionality over repeated iterations.
Why use it?
It turns a broad feature idea into a written list of capabilities and a repeatable development process. It also detects the application's language, build system, and project type so the generated workflow fits the codebase.

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/aminry/auto-claude-plugins/create
Clone the repo
git clone --depth 1 https://github.com/aminry/auto-claude-plugins
Per session 14 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 7,601 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 1 finding. 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 $0.00014 $0.07601
Opus 5 $0.00007 $0.03800
Sonnet 5 $0.00003 $0.01520
Haiku 4.5 $0.00001 $0.00760

Measured yesterday against content hash bb44685dfa59, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

create scanned grade A with 1 finding 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 yesterday.

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.

Makes network callslowCapability

Not a fault in itself. Listed so you know the mod talks to something, and to what.

| Web backend API | Start server + curl endpoint, or run integration test targeting new functionality |
plugins/auto-develop/commands/create.md · 865 lines

How it starts

The opening of the file, as written. The whole thing — 865 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Create a custom auto-develop skill for a specific feature in the application. Analyzes the feature, interviews the user to discover and curate new capabilities, builds a capability specification, instruments code if needed, and generates a complete auto-develop skill that iteratively implements new functionality.

The feature to develop: $ARGUMENTS


Overview

This skill creates two files:

  1. A capability specification (.claude/commands/spec-{feature}.md) — the enumerated list of new capabilities to implement
  2. An auto-develop skill (.claude/commands/auto-develop-{feature}.md) — the iterative development loop

It does this through a structured process that minimizes back-and-forth — all user input is gathered in a single round after initial analysis.


Phase 0: App & Stack Detection

Before analyzing the feature, discover the project's technology stack and app type. This "stack profile" informs all later phases.

0a: Language & Build System

Scan the project root for manifest files and infer the stack:

Manifest File Language/Ecosystem
package.json Node.js / TypeScript / JavaScript
pyproject.toml, setup.py, requirements.txt Python
go.mod Go
Cargo.toml Rust
Gemfile Ruby
build.gradle, pom.xml Java / Kotlin
Makefile (alone) C / C++ / generic
mix.exs Elixir
Package.swift Swift
*.csproj, *.sln C# / .NET

Also detect:

  • Package manager: npm/yarn/pnpm, pip/poetry/uv, cargo, go modules, bundler, maven/gradle, etc.
  • Framework: Next.js, Django, Flask, FastAPI, Rails, Gin, Actix, Spring Boot, Phoenix, etc.
  • Test runner: vitest/jest/mocha, pytest, go test, cargo test, rspec, junit, etc.
  • Linter/type checker: eslint/tsc, ruff/mypy/pylint, go vet, cargo clippy, rubocop, etc.
  • Build command: npm run build, python -m build, go build, cargo build, make, etc.

0b: App Type Classification

Based on the code structure, classify the app type:

Read the full file on GitHub · 865 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. yesterday First seen · 865 lines · 14 tokens per session scan A bb44685dfa59

Subscribe to this mod's changes

create is a command published in the GitHub repository aminry/auto-claude-plugins (1 stars, last pushed 3mo ago), licensed MIT. It adds 14 tokens to every session and 7,601 once invoked, about $0.0001 per session on Opus 5. A static security scan graded it A with 1 finding (makes network calls). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.