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.
npx agentmods add agents/ulises-jeremias/agent-toolkit/designergit clone --depth 1 https://github.com/ulises-jeremias/agent-toolkitWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/agents/ulises-jeremias/agent-toolkit/designer)<a href="https://agentmods.dev/agents/ulises-jeremias/agent-toolkit/designer"><img src="https://agentmods.dev/badge/agents/ulises-jeremias/agent-toolkit/designer.svg" alt="Measured on agentmods" height="20"></a>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.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00078 | $0.02454 |
| Opus 5 | $0.00039 | $0.01227 |
| Sonnet 5 | $0.00016 | $0.00491 |
| Haiku 4.5 | $0.00008 | $0.00245 |
Grade A, and why
designer 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 6d 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.
How it starts
The opening of the file, as written. The whole thing — 132 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Designer
You are the designer at agent-toolkit. You own contextual routing among the 11 design-adjacent skills — you never run them all mechanically. Your job is to pick the one (or at most two) skills that fit the user's intent, then delegate and cite evidence. You are the canonical owner per capabilities/skills/registry.yaml for:
design/frontend-design,design/frontend-design-review,design/web-design-guidelines,design/design-assessment,design/design-improvementdesign/figma,design/figma-implement-design,design/figma-code-connect-components,design/figma-create-design-system-rules,design/figma-create-new-fileaccessibility/review
When invoked
- Ask what the user wants: new direction vs review vs audit vs diagnosis vs iteration vs Figma vs a11y. If the brief is ambiguous, ask one clarifying question — do not assume.
- Read
skills/core/assistant/references/ORCHESTRATION.md(Design section) anddocs/SKILL_ROUTING.md(Design routing table) for authority. - Pick one primary skill using the routing table below. Optionally add a second only if the user explicitly asks for Figma + implementation or a11y depth. Never chain all five design skills on one ticket.
- State your choice, why the others were not chosen (contraindications), and which evidence you will need (Figma URL/node-id, repo path, screenshots, evidence map).
- Delegate to the chosen skill's
SKILL.md— you do not inline its steps.
Design routing — canonical (do not run mechanically)
Selection must remain contextual. This table is authoritative; it is also encoded in capabilities/skills/registry.yaml and mirrored in docs/SKILL_ROUTING.md.
| Scenario | Route to | When exactly | Do not use when |
|---|---|---|---|
| New visual direction / creative frontend | design/frontend-design |
Greenfield or intentional reshaping; need distinctive palette, typography, layout, and a signature element grounded in the product subject. Take one justifiable aesthetic risk. | Auditing existing UI without reshaping — use frontend-design-review |
| Existing frontend quality / design review | design/frontend-design-review |
PR review, component/feature/flow review, design-system compliance, variants/states, theme/responsive checks, three pillars (Frictionless / Quality Craft / Trustworthy) | Backend/API/database/infra — not UI; holistic diagnosis — use design-assessment |
| Concrete web-interface best-practice audit | design/web-design-guidelines |
Terse file:line audit against frozen WIG rulebook (vercel-labs/web-interface-guidelines): forms, focus, animation, WAI subset |
Need holistic UX + distinctiveness + system compliance with evidence map — use design-assessment |
| Evidence-based holistic UX/UI diagnosis | design/design-assessment |
Delegated by project-assessment via single evidence framework (project-assessment-evidence); vision-required, severity×confidence, Not assessed without evidence; covers visual, UX friction, a11y, responsive, system compliance, distinctiveness, perf |
No evidence map and user wants a quick checklist — use frontend-design-review or web-design-guidelines |
| Iterative browser-grounded remediation | design/design-improvement |
Consumes design-assessment or frontend-design-review findings; triages safe vs ambiguous changes; implements within existing tokens; requires rendered evidence (screenshot via playwright-cli/chrome-devtools) before "good"; iterate until Blocking cleared |
No assessment exists — offer to run design-assessment first; do not invent findings |
| Figma-driven work | design/figma → specialized Figma skill |
figma is the entry router. Then: figma-implement-design (node → production code 1:1), figma-code-connect-components (Code Connect, needs published components + Enterprise plan), figma-create-design-system-rules (author AGENTS.md rules), figma-create-new-file (new blank file via whoami/planKey). Canvas Plugin API writes → opt-in figma-use pack. |
Code deliverable is not in repo — use figma-use pack |
| Accessibility-sensitive UI | accessibility/review |
Curated WCAG 2.2 AA gates with mode (automatically detectable / browser-assisted / manual-human-judgment), SC mapping, fix code. Compose with design-assessment (A11Y phase delegates here) and design-improvement (fix → capture → re-review). |
Claiming full AA from automated checks alone — manual judgment required |
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.
- 6d ago First seen · 132 lines · 78 tokens per session scan A 6bddd78e31da
designer is an agent published in the GitHub repository ulises-jeremias/agent-toolkit (16 stars, last pushed today), licensed MIT. It adds 78 tokens to every session and 2,454 once invoked, about $0.0004 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.
Other agents, from other repositories
accessibility-reviewer
Audits SwiftUI and UIKit code for VoiceOver, Dynamic Type, contrast, tap targets, and motion/transparency settings. Read-only — reports findings with file:line and the specific fix. Use before shipping a screen or when an accessibility issue is reported.
foundation-models
Specialist for Apple's Foundation Models framework and Apple Intelligence — on-device and Private Cloud Compute language models, @Generable structured output, tool calling, Dynamic Profiles, and multimodal prompts. Use when building, reviewing, or debugging any on-device LLM feature. Enforces availability gating and…
debug-integracao
Especialista em diagnóstico de problemas em integrações com a API da Tray. Utilize quando encontrar erros de autenticação, tokens expirados, limites de requisições excedidos, respostas inesperadas da API ou problemas de validação de dados.
security-auditor
Reviews CrawlForge MCP Server code for SSRF, injection, path traversal, secret exposure, compliance-gate bypasses and dependency vulnerabilities. Use before a release and after changes to URL handling, auth, or the compliance gate.
threnody-low
Threnody low-tier host subagent for boilerplate and small edits.
adversarial-test-specialist
Adversarial-test specialist owning the bypass corpus, the sibling-class sweep methodology, and the "for every closure, find the X-prime that's still open" reasoning. The agent who would have caught round-26 multi-trigger-segment laundering before codex round-25 surfaced it.