specify

A documentation tool that turns design decisions into detailed instructions for developers and other teams. It can organize handoff materials, assets, test plans, and related decisions.

In plain words
What is it for?
Use it to write design specifications, prepare engineering handoffs, document decisions, and plan design reviews or testing.
Why use it?
It reduces misunderstandings between design and engineering and helps prevent rework during implementation.

Cursor rule for Cursor

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 rules/ghaida/intent/specify
Clone the repo
git clone --depth 1 https://github.com/ghaida/intent

Made for: Cursor.

Per session 99 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 3,238 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 $0.00099 $0.03238
Opus 5 $0.00049 $0.01619
Sonnet 5 $0.00020 $0.00648
Haiku 4.5 $0.00010 $0.00324

Measured 2d ago against content hash 8449b4f71ab9, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

specify 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.

.cursor/rules/specify.mdc · 318 lines

How it starts

The opening of the file, as written. The whole thing — 318 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Specify — Bridge Design to Engineering

Overview

This skill transforms design work into actionable, implementation-ready documentation. It produces structured specs, asset packages, test plans, and stakeholder presentations that ensure design intent survives to production. Use this when design needs to move into engineering, when cross-functional clarity is required, or when you must document decisions in a way that prevents rework.


Skill family

Specify works alongside the full Intent skill system:

  • /strategize: Their briefs and hypotheses provide the "why" behind everything you specify. Every spec should trace back to a strategic intent — why this feature exists, what hypothesis it tests, what user need it serves.
  • /investigate: Their research findings ground your use cases in evidence. Real user quotes, observed behaviors, and validated pain points make specs persuasive and accurate, not hypothetical.
  • /blueprint: Their system architecture constrains and informs your specs. Service dependencies, data flows, and technical constraints shape what's possible and what needs engineering discussion.
  • /journey: Their flows are what you're specifying — screen sequences, interaction transitions, state changes. Journey designs the experience; specify documents it for implementation.
  • /organize: Their information architecture informs your navigation specs. Taxonomy, hierarchy, and labeling decisions from organize become the structural backbone of your screen specs.
  • /articulate: Their copy work feeds directly into your copy matrices. Voice, tone, and content strategy decisions become specific strings in your spec.
  • /fortify: Their edge case analysis becomes part of your spec. Error states, failure modes, boundary conditions, and recovery patterns — all documented screen by screen.
  • /include: Their accessibility requirements go into every screen spec. ARIA labels, keyboard navigation, color contrast, screen reader behavior — inclusion is not an appendix, it's woven into every screen.
  • /evaluate: Their assessment identifies gaps in your specs. Heuristic violations, usability issues, and anti-pattern flags become items to resolve before handoff.
  • /measure: Their success metrics define your test plan criteria. Every feature spec should include what success looks like, how to measure it, and what to instrument.
  • /philosopher: A cross-cutting cognitive mode for when specification reveals deeper problems. Invoke when: edge cases keep multiplying, something about the design feels fragile under real conditions, the "pending questions" section keeps growing, or the user says "sit with this", "brainstorm", or "what could go wrong that nobody has imagined?" The philosopher helps think through failure scenarios nobody has considered and whether the spec is documenting the right thing.

Read the full file on GitHub · 318 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 · 318 lines · 3,238 tokens per session scan A 8449b4f71ab9

Subscribe to this mod's changes

specify is a cursor rule published in the GitHub repository ghaida/intent (139 stars, last pushed 1mo ago), licensed CC0-1.0. It adds 99 tokens to every session and 3,238 once invoked, about $0.0005 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.