giselle: Agent for Claude Code

.claude/agents/ux-guardian.md

ux-guardian is an agent for Claude Code from toyamarinyon/giselle. It costs 43 tokens per session (768 once invoked), scanned A, a copy of ux-guardian, Apache-2.0.

An agent that finds test cases by considering the experience and expectations of real users, not only whether functions and APIs work. It examines edge cases, unexpected action sequences, long-term use, errors, partial success, and concurrent activity.

In plain words
What is it for?
Reviewing code changes, discovering overlooked tests, checking recovery from errors, and testing behaviour at unusual inputs or timings.
Why use it?
It helps uncover user-facing failures that ordinary tests may miss, especially when behaviour is technically valid but surprising or frustrating.

Agent for Claude Code

Written for Claude Code: installed under .claude/.

This is toyamarinyon/giselle's own configuration. It tells Claude Code how to work on giselle itself, so it is not a mod to install elsewhere. Copy it as a starting point and replace the rules that are about this project. Everything giselle configures →

Reuse

Borrowing it

Nothing to install: this file belongs to toyamarinyon/giselle. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.

Copy the file
curl -O https://raw.githubusercontent.com/toyamarinyon/giselle/main/.claude/agents/ux-guardian.md
Clone the repo
git clone --depth 1 https://github.com/toyamarinyon/giselle

Made for: Claude Code.

Wrote 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.

agentmods badge for ux-guardian

README.md
[![agentmods](https://agentmods.dev/badge/agents/toyamarinyon/giselle/ux-guardian/github.svg)](https://agentmods.dev/agents/toyamarinyon/giselle/ux-guardian)
Your own site
<a href="https://agentmods.dev/agents/toyamarinyon/giselle/ux-guardian"><img src="https://agentmods.dev/badge/agents/toyamarinyon/giselle/ux-guardian/github.svg" alt="Measured on agentmods" height="20"></a>

Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.

agentmods 80×15 button for ux-guardian

Your own site · 80×15
<a href="https://agentmods.dev/agents/toyamarinyon/giselle/ux-guardian"><img src="https://agentmods.dev/badge/agents/toyamarinyon/giselle/ux-guardian.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 43 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 768 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. A grade says what 26 rules found in the file — not that it is safe.
Origin 100% copy Near-identical to another mod 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 5.1 $0.00043 $0.00768
Opus 5 $0.00022 $0.00384
Sonnet 5 $0.00009 $0.00154
Haiku 4.5 $0.00004 $0.00077

Measured 11d ago against content hash 1cef1df4adff, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-10, from the pricing page.

Security

Grade A, and why

ux-guardian 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 11d 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.

Origin

This is a copy

100% identical to ux-guardian — 0 lines differ, which has more behind it and is treated as the original. This page carries a canonical link to it rather than competing with it.

.claude/agents/ux-guardian.md · 121 lines

How it starts

The opening of the file, as written. The whole thing — 121 lines — stays where its author put it; the contents beside it link to each section on GitHub.

You are a guardian of user experience. When you look at code, you don't just verify functions or APIs work—you deeply understand what they mean to users and what expectations they create.

Thinking Behind Test Discovery

Why Does This Code Exist?

Before enumerating test cases, first understand the problem the application is trying to solve. In a node-based visual programming environment like Giselle:

  • Users seek creative expression
  • They want to bring ideas to life without being bothered by technical details
  • They expect immediate feedback and predictable behavior

The moments when these expectations are betrayed are the most important test cases.

Seeing the Invisible

Great test cases go beyond obvious functional tests, holding these perspectives:

Experience Across Time

  • Intuitiveness on first touch
  • Consistency through repeated use
  • Performance degradation after long-term use

Behavior at the Edges

  • Extremely small/large inputs
  • Operations in unexpected sequences
  • Concurrent execution and race conditions

Betrayed Expectations

  • What users think "should happen"
  • Graceful recovery from errors
  • Handling partial successes

Framework for Thinking

1. Start with User Journeys

When reading the codebase, first imagine the paths users traverse:

First visit → Exploration → First success → Mastery → Advanced use → Error encounter → Recovery

Consider what should happen and what must not happen at each stage.

2. Follow the Web of Interactions

In node-based systems like Giselle, individual nodes may appear independent but actually exist within a complex web of interactions:

  • Data flow between nodes
  • Execution order dependencies
  • Resource sharing and contention
  • State propagation and synchronization

These interaction boundaries are the most fragile places needing tests.

3. Question the "Obvious"

The parts developers assume "obviously work" often hide important test cases:

  • "Cancel button always works" → Really? Even during processing?
  • "Save will succeed" → What if storage is full?
  • "AI returns responses" → What about timeouts? Rate limits?

Read the full file on GitHub · 121 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. 11d ago First seen · 121 lines · 43 tokens per session scan A 1cef1df4adff

Subscribe to this mod's changes

ux-guardian is an agent published in the GitHub repository toyamarinyon/giselle (4 stars, last pushed 4mo ago), licensed Apache-2.0. It adds 43 tokens to every session and 768 once invoked, about $0.0002 per session on Opus 5. A static security scan graded it A with 0 findings. It is 100% identical to ux-guardian, differing in 0 lines, and is treated as a copy.

Related

Other agents, from other repositories

ring:qa-frontend

Senior Frontend QA Analyst for React/Next.js. Supports 5 modes — unit (default), accessibility, visual, e2e, performance. Dispatched with mode parameter; loads mode-specific file from qa-frontend-modes/.

LerianStudio/ring · 53 tokens

accessibility_specialist

Accessibility specialist for WCAG compliance auditing, ARIA implementation review, keyboard navigation testing, and inclusive design assessment. Use when the task requires accessibility audits, screen reader compatibility checks, color contrast verification, or ARIA role validation. For example: auditing a web app for…

josstei/maestro-orchestrate · 83 tokens

a11y-write-playwright-tests

Write Playwright accessibility tests for a component or page: axe scan file and keyboard-only variant. Triggers when the user asks to 'add a11y tests', 'write axe test', 'add keyboard test', or 'create accessibility test for X'.

AmadeusITGroup/otter · 59 tokens

ia-accessibility-tester

WCAG 2.1/2.2 accessibility audit: keyboard navigation, screen reader, contrast, ARIA, forms, cognitive. Use for accessibility review, WCAG compliance, or inclusive design assessment.

iliaal/whetstone · 48 tokens

product

You are the product quality evaluator for the Kernel AI platform. You think like a user, not an engineer.

isaacsight/kernel · 0 tokens

visual-qa-agent

Visual regression testing and design comparison agent. Renders web app pages via Chrome DevTools and Playwright, captures Figma designs, and produces structured visual diff reports with cross-browser testing.

PMDevSolutions/Aurelius · 42 tokens