create

A guided feature-planning command that examines a codebase, asks informed questions, compares implementation approaches, and produces context, requirements, and task files.

In plain words
What is it for?
Use it to explore an unfamiliar repository and prepare a structured specification for a new feature, including implementation tasks.
Why use it?
It reduces the manual work of turning a feature idea into a plan grounded in the existing code. A review step helps check the specification before it is finalized.

Command for Claude Code

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/marcelsud/spec-driven-agentic-development/create
Clone the repo
git clone --depth 1 https://github.com/marcelsud/spec-driven-agentic-development

Made for: Claude Code.

Per session 16 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 3,246 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 $0.00016 $0.03246
Opus 5 $0.00008 $0.01623
Sonnet 5 $0.00003 $0.00649
Haiku 4.5 $0.00002 $0.00325

Measured 2d ago against content hash e80f54f3de54, 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 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/spec/create.md · 517 lines

How it starts

The opening of the file, as written. The whole thing — 517 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Unified Feature Specification (Enhanced)

Generate complete feature specification with deep codebase exploration using Task agents.

Your Task

Create a complete specification for: $ARGUMENTS

Enhanced Flow Overview

  1. Triage - Assess complexity, determine exploration depth
  2. Explore - Spawn Explore agents to understand codebase (parallel)
  3. Initial Questions - Gather context informed by exploration
  4. Plan - Spawn Plan agents for approaches and strategies (parallel)
  5. Synthesize - Create exploration.md with all findings
  6. Generate - Create context.md, requirements.md, tasks.md
  7. Review - Run /spec:review to validate spec before finalizing
  8. Summary - Present all generated files

Phase 0: Triage

Quickly assess feature complexity to determine how many Explore agents to spawn.

Assessment Steps:

  1. Parse the feature description for complexity keywords
  2. Check if features/ directory exists with established patterns: Glob("features/*/context.md")
  3. Estimate project size: Glob("**/*.{ts,js,py,go,rs,java}") (count source files)

Complexity Matrix:

Indicator Low Medium High
Description keywords Simple CRUD, basic Auth, API, data Integration, realtime, complex
Existing features 0 1-3 4+
Source files < 20 20-100 > 100

Agent Count Decision:

  • Low complexity: 1 Explore agent (comprehensive)
  • Medium complexity: 2 Explore agents (patterns + similar features)
  • High complexity: 3 Explore agents (patterns, similar features, tech stack)

Phase 1: Explore Agents

Use the Task tool with subagent_type: "Explore" to spawn codebase exploration agents. Run all applicable agents in parallel.

Agent 1: Patterns & Conventions (Always spawn)

Task tool parameters:
- subagent_type: "Explore"
- description: "Explore codebase patterns"
- prompt: |
    Explore the codebase to understand existing patterns for implementing: [feature-description]

    Search for and document:
    1. Directory structure - Where does similar code live? (use Glob)
    2. Naming conventions - File names, function names, class names
    3. Code organization - How are modules/services/controllers structured?
    4. Error handling - How are errors caught, logged, and reported?
    5. Configuration - How is config managed? Environment variables?

    Use Glob and Grep to find concrete examples. Return findings with specific file paths.

Read the full file on GitHub · 517 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. 2d ago First seen · 517 lines · 16 tokens per session scan A e80f54f3de54

Subscribe to this mod's changes

create is a command published in the GitHub repository marcelsud/spec-driven-agentic-development (174 stars, last pushed 9mo ago), licensed MIT. It adds 16 tokens to every session and 3,246 once invoked, about $0.0001 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.