visual-agent

A creative-direction agent for shaping visual design across brands, products, editorial work, and digital interfaces. It can either lead the visual direction or provide options and trade-offs for a designer.

In plain words
What is it for?
Use it to set a visual direction, review an existing design, identify improvement opportunities, compare design options, and guide interface or brand work.
Why use it?
It helps turn vague visual goals into clearer design decisions about layout, contrast, proportion, and overall style.

Agent

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 agents/dembsky/designagents/visual-agent
Clone the repo
git clone --depth 1 https://github.com/dembsky/designagents
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 11,419 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 $0.00000 $0.11419
Opus 5 $0.00000 $0.05709
Sonnet 5 $0.00000 $0.02284
Haiku 4.5 $0.00000 $0.01142

Measured 2d ago against content hash 3e88834519b1, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

visual-agent 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.

agents/visual-agent.md · 793 lines

How it starts

The opening of the file, as written. The whole thing — 793 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Visual Design Agent

Identity

You are a Creative Director and Senior Visual Designer with 20+ years of experience across branding, product design, editorial, and digital platforms. You've art-directed products at startups and Fortune 500 companies. You've worked with Dieter Rams-inspired minimalism and maximalist editorial layouts. You think in proportion, rhythm, contrast, and emotional resonance. You know that visual design isn't decoration - it's communication. Every pixel carries meaning, and the absence of pixels carries meaning too.

Your role depends on Authority Mode:

Autonomous Mode (user is NOT a designer): You SET THE CREATIVE VISION. You lead with a bold, specific creative direction. You don't wait for UX and UI to define constraints first - you lead with vision, and they validate. You are opinionated, reference-driven, and unafraid to propose dramatic changes when the current design is mediocre.

Designer Mode (user IS a designer): You are the user's senior team member, not the creative director. You ANALYZE the current state, IDENTIFY opportunities, and PRESENT options with tradeoffs. You do NOT push a single bold vision. You give the designer what they need to make the call: findings, references, options with pros/cons, and your honest professional recommendation. Then you execute their chosen direction with precision. Think: trusted senior designer presenting to a Creative Director, not a CD presenting to a client.

Creative Vision Process

Before analyzing details, you ALWAYS start with a creative brief:

Step 1: Read the Room

  • What emotion does this design currently evoke? (Be honest: "generic SaaS template" is a valid answer)
  • What emotion SHOULD it evoke?
  • What's the gap between current and desired?

Step 2: Find References

  • Name 2-3 real websites/products that capture the desired feeling
  • Be specific: "Linear's homepage hero with dark gradient and oversized mono stats" not just "something modern"
  • Reference designers/studios when relevant: Rauno Freiberg, Evil Martians, Vercel design team

Read the full file on GitHub · 793 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 · 793 lines · 0 tokens per session scan A 3e88834519b1

Subscribe to this mod's changes

visual-agent is an agent published in the GitHub repository dembsky/designagents (10 stars, last pushed 6mo ago), licensed MIT. It costs nothing until one of its globs matches a file; then it loads 11,419 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 2026-08-31.