A command for expanding an existing GUI test harness, which is a set of tools and definitions for testing software through its graphical interface. It reviews the harness and the source code to find missing coverage.

Use it to inspect a local software project, compare its capabilities with the existing GUI harness, and identify missing actions, incomplete forms, or uncovered feature groups.

It helps reveal software actions, form fields, or workflows that the current GUI tests do not cover. A focus area can limit the review to one part of the software.

refine — Command for coding agents by ImL1s: A command for expanding an existing GUI test harness…

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/iml1s/gui-anything/refine

Clone the repo

git clone --depth 1 https://github.com/ImL1s/GUI-Anything

Or install gui-anything, the plugin that ships this one along with the rest of its 5 commands.

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.

README.md

[![agentmods](https://agentmods.dev/badge/commands/iml1s/gui-anything/refine.svg)](https://agentmods.dev/commands/iml1s/gui-anything/refine)

Your own site

<a href="https://agentmods.dev/commands/iml1s/gui-anything/refine"><img src="https://agentmods.dev/badge/commands/iml1s/gui-anything/refine.svg" alt="Measured on agentmods" height="20"></a>

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 469 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.

Model	Per session	Once invoked
Fable 5	$0.00000	$0.00469
Opus 5	$0.00000	$0.00234
Sonnet 5	$0.00000	$0.00094
Haiku 4.5	$0.00000	$0.00047

Measured 3d ago against content hash ce43cce9c6e3, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

refine 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 3d 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.

gui-anything-plugin/commands/refine.md · 71 lines

How it starts

The opening of the file, as written. The whole thing — 71 lines — stays where its author put it; the contents beside it link to each section on GitHub.

gui-anything:refine Command

Expand and improve an existing GUI harness with additional coverage.

CRITICAL: Read HARNESS.md First

Before refining, read ./HARNESS.md. It defines the standards that all refinements must follow.

Usage

/gui-anything:refine <software-path> [focus]

Arguments

<software-path> - Required. Local path to the software source code.
[focus] - Optional. Specific area to focus refinement on (e.g., "export workflows", "settings panel", "batch operations").

What This Command Does

Step 1: Inventory Current Coverage

Read the existing gui-schema.json to catalog all current actions
Read the existing mcp-tools.json to catalog all MCP tools
Count components, actions, and tool definitions

Step 2: Analyze Software Capabilities

Re-analyze the source code for functions not yet mapped
Identify missing parameter types or widget optimizations
Find uncovered action groups

Step 3: Gap Analysis

Compare current GUI coverage against full software capabilities
If [focus] is provided, narrow the gap analysis to that area
List missing actions, incomplete forms, and uncovered features
Prioritize gaps by user impact

Step 4: Implement New Components

Add new actions to gui-schema.json
Add corresponding form components
Update sidebar navigation if new groups are added
Add matching MCP tools to mcp-tools.json

Step 5: Expand Tests

Add tests for new schema entries
Add tests for new MCP tools
Update TEST.md with new test plan entries

Step 6: Update Documentation

Update README.md with new features
Update <SOFTWARE>.md with new analysis findings

Example

# Broad gap analysis and expansion
/gui-anything:refine ./calculator

# Focus on specific area
/gui-anything:refine ./image-editor "batch processing and export presets"

Success Criteria

New actions added to schema and working
MCP tools match all actions
All existing + new tests pass
Documentation updated

Read the full file on GitHub · 71 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.

3d ago First seen · 71 lines · 0 tokens per session scan A ce43cce9c6e3

Subscribe to this mod's changes

refine is a command published in the GitHub repository ImL1s/GUI-Anything (5 stars, last pushed 5mo ago), licensed Apache-2.0. It costs nothing until one of its globs matches a file; then it loads 469 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.

E2E testing Testing ai-agent cross-platform gui mcp plugin

refine