define

Capture and validate requirements in one pass (Phase 1).

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/luanmorenommaciel/agentspec/define
Clone the repo
git clone --depth 1 https://github.com/luanmorenommaciel/agentspec

Made for: Claude Code.

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 1,109 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 $0.00014 $0.01109
Opus 5 $0.00007 $0.00554
Sonnet 5 $0.00003 $0.00222
Haiku 4.5 $0.00001 $0.00111

Measured yesterday against content hash c8c152fda3fb, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

define 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 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.

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/workflow/define.md · 113 lines

How it starts

The opening of the file, as written. The whole thing — 113 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Define Command

Capture requirements and validate them in one pass (Phase 1)

Usage

/define <input> [--judge[=MODE]]

Examples

# From a BRAINSTORM document (recommended after /brainstorm)
/define .claude/sdd/features/BRAINSTORM_NOTIFICATION_SYSTEM.md

# From meeting notes or raw input
/define notes/meeting-notes.md
/define "Build an API gateway for user management"
/define docs/stakeholder-email.txt

# With cross-model judge for spec quality verification (opt-in)
/define BRAINSTORM_AUTH.md --judge                  # advisory, default openai/gpt-4o
/define BRAINSTORM_AUTH.md --judge=strict           # gated — FAIL blocks completion
/define BRAINSTORM_AUTH.md --judge=anthropic/claude-opus-4  # custom model (advisory)
/define BRAINSTORM_AUTH.md --judge=strict:openai/gpt-4o     # gated + custom model

What happens

This command is the Phase 1 entrypoint; the methodology — process steps, scoring rubric, quality gate, anti-patterns — lives in the skill, not here.

  1. Load .claude/skills/sdd-define/SKILL.md and follow it: input classification, entity extraction, technical and data engineering context, clarity scoring against the 12/15 gate, gap filling.
  2. Produce the DEFINE document at .claude/sdd/features/DEFINE_{FEATURE_NAME}.md (structure: .claude/sdd/templates/DEFINE_TEMPLATE.md) and update document statuses per the workflow contracts.
  3. Suggest the next phase: /design .claude/sdd/features/DEFINE_{FEATURE_NAME}.md.

Optional Judge Pass (--judge)

Runs only if the user invoked with --judge[=MODE]. This is a cross-model second opinion on the spec you just wrote, using a non-Claude model via OpenRouter. Defaults are designed so most users never notice the flag exists.

Flag parsing (parse from the user's command args):

Input Mode Model
--judge advisory phase default (openai/gpt-4o for define)
--judge=strict gated phase default
--judge=MODEL_SLUG advisory MODEL_SLUG
--judge=strict:MODEL_SLUG gated MODEL_SLUG

Read the full file on GitHub · 113 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 · 113 lines · 14 tokens per session scan A c8c152fda3fb

Subscribe to this mod's changes

define is a command published in the GitHub repository luanmorenommaciel/agentspec (244 stars, last pushed 2d ago), licensed MIT. It adds 14 tokens to every session and 1,109 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 2026-09-01.