Skill flagged — suspicious patterns detected

ClawHub Security flagged this skill as suspicious. Review the scan results before using.

GT7 Tuner — Gran Turismo 7 AI Tuning Assistant

v1.0.0

AI tuning assistant for Gran Turismo 7. Send a photo of your GT7 settings screen → get a complete optimized tune via GT Pro Tune calculator. Supports all 581...

0· 111·0 current·0 all-time

Install

OpenClaw Prompt Flow

Install with OpenClaw

Best for remote or guided setup. Copy the exact prompt, then paste it into OpenClaw for iamstevebot/gt7-tuner.

Previewing Install & Setup.
Prompt PreviewInstall & Setup
Install the skill "GT7 Tuner — Gran Turismo 7 AI Tuning Assistant" (iamstevebot/gt7-tuner) from ClawHub.
Skill page: https://clawhub.ai/iamstevebot/gt7-tuner
Keep the work scoped to this skill only.
After install, inspect the skill metadata and help me finish setup.
Use only the metadata you can verify from ClawHub; do not invent missing requirements.
Ask before making any broader environment changes.

Command Line

CLI Commands

Use the direct CLI path if you want to install manually and keep every step visible.

OpenClaw CLI

Bare skill slug

openclaw skills install gt7-tuner

ClawHub CLI

Package manager switcher

npx clawhub@latest install gt7-tuner
Security Scan
VirusTotalVirusTotal
Suspicious
View report →
OpenClawOpenClaw
Benign
high confidence
Purpose & Capability
Name/description (GT7 tuning assistant) match the instructions: the SKILL.md describes reading a screenshot, extracting values with vision/OCR, and automating GT Pro Tune's web app to retrieve a tune. Required resources (none) are proportionate to that purpose.
Instruction Scope
Instructions focus on OCR, unit conversion, user confirmation, and browser automation of https://app.gtprotune.com. They include in-page JavaScript eval snippets to select typeahead items and click buttons — this is necessary for web automation but does execute arbitrary JS in the page context. The doc advises not to log credentials but does not define how or where credentials should be stored by the agent.
Install Mechanism
Instruction-only skill with no install steps or third-party downloads. No code files to write or arbitrary network installs are requested.
Credentials
No environment variables, keys, or unrelated credentials are requested. The only external account referenced is the GT Pro Tune account (appropriate for the described integration).
Persistence & Privilege
Skill is not forcing permanent inclusion (always: false) and does not request system-wide configuration changes or access to other skills' credentials. The skill can be invoked by the agent (default), which is expected for this kind of automation.
Assessment
This skill appears to do what it says: read GT7 screenshots and drive the GT Pro Tune web app to produce setups. Before using it: (1) Do not paste your GT Pro Tune password into chat — prefer entering credentials only when the agent's browser automation prompts a secure login flow. (2) Consider testing with a throwaway/free GT Pro Tune account if you are uneasy about automation interaction. (3) Be aware the skill runs JavaScript in the GT Pro Tune page (eval/click); that's normal for web automation but means the agent will execute code in the page context. (4) If you require guarantees about credential storage or retention, ask the skill author how credentials are handled (the SKILL.md advises not to log them but gives no storage details). Otherwise this skill is internally consistent with its purpose.

Like a lobster shell, security has layers — review code before you run it.

Runtime requirements

🏎️ Clawdis
latestvk971b8zbe10yvts59d1gxp79n9839h2e
111downloads
0stars
1versions
Updated 1mo ago
v1.0.0
MIT-0

GT7 Tuner — Gran Turismo 7 AI Tuning Assistant

Turn any GT7 car into a weapon. Send a screenshot, get a pro tune.

Prerequisites

⚠️ You need a GT Pro Tune account (https://app.gtprotune.com/)

  • Free accounts have limited features
  • Pro account recommended ($4.99/month) — unlocks full suspension/transmission calculations, Image Analyzer, and Race Engineer
  • Sign up at https://app.gtprotune.com/register
  • Store your credentials securely (never log or echo them)

When to Use

  • User wants to tune a car in Gran Turismo 7
  • User sends a screenshot/photo of their GT7 car settings
  • User mentions GT Pro Tune, car setup, suspension, gearing, or tuning
  • User says something like "tune my [car] for [track]"
  • Any Gran Turismo 7 tuning-related request

How It Works

The Quick Flow

  1. User sends a photo of their GT7 settings screen (from TV/monitor)
  2. Agent reads the values using vision (OCR) — PP, weight, HP, tire compound, etc.
  3. Agent confirms values with user (confidence check — only use values that are clearly readable)
  4. Agent fills GT Pro Tune via agent-browser automation
  5. Agent reads the calculated tune and reports it back
  6. User applies the tune in GT7

Photo OCR Pipeline

When the user sends a GT7 screenshot (photo of TV screen):

  1. Read the image with vision capabilities
  2. Extract all visible values: PP, weight (kg or lbs), HP, torque, tire compound, etc.
  3. For each value, assess confidence:
    • High confidence (>95%): use directly
    • Low confidence: ask user to confirm or re-send a clearer photo
  4. Convert units if needed (kg → lbs, kW → HP, Nm → ft-lb)
  5. Present extracted values to user for confirmation before proceeding

Tips for better photos:

  • Take the photo straight-on (reduce angle)
  • Minimize screen glare/reflections
  • Ensure the full settings screen is visible
  • PS5 screenshots via Share button → PS App are ideal (pixel-perfect)

GT Pro Tune Web App

Login

agent-browser open "https://app.gtprotune.com/login"
# Fill email and password fields, click Login
# Credentials must be stored securely — NEVER log, echo, or display them

Important: Angular Input Handling

GT Pro Tune is built with Angular 17. When filling form fields:

  • Use agent-browser fill @ref "value" followed by agent-browser press "Tab" — this triggers Angular's change detection
  • Do NOT use eval/DOM manipulation to set values — Angular's reactive forms won't detect the changes
  • For dropdowns (ngb-typeahead): type the search text, wait for results, then use eval to click the matching option
  • For select elements: use agent-browser select @ref "value"

App Structure

Tab 1: Car/Track Settings

FieldTypeNotes
VehicleTypeahead search581 cars, type to filter
Performance PointsNumberPP rating
WeightNumberIn pounds (lbs)
TrackTypeahead searchType to filter
Tire SizeNumberIn inches
Front TiresDropdownComfort/Sport/Racing × Hard/Medium/Soft
Rear TiresDropdownSame options as front

Tab 2: Suspension Settings

SectionFields
Base PerformanceFront-Rear Weight Balance (%)
StabilityLow Speed, High Speed (text: understeer/oversteer)
Rotational GG at 40mph, 75mph, 150mph
SuspensionBody Height Front/Rear (mm), Oversteer/Understeer slider
AerodynamicsDownforce Front/Rear
Race/Track ParametersSpring Stiffness, ARB Stiffness, Tire Wear multiplier
Other AdjustmentsCorner Entry/Exit balance, Offset Spring Frequency

Tab 3: Transmission Settings

SectionFields
Engine DataHP, RPM (min/max/peak), Max Torque (ft-lb)
DifferentialLSD Setting
GearsTop Speed (MPH), Number of Gears (3-9), Final Gear, Gear Ratios
Corner SpeedMin Corner Speed (MPH), Min Corner Speed Gear

Vehicle Selection (Angular Typeahead)

agent-browser click @e15                    # Focus vehicle search
agent-browser keyboard type "F8 Tributo"    # Type car name (spaces may cause CDP errors — use partial names)
sleep 2                                     # Wait for dropdown
agent-browser eval "(function(){var items=document.querySelectorAll('ngb-typeahead-window button');var match=Array.from(items).find(function(e){return e.textContent.includes('F8 Tributo')});if(match){match.click();return 'selected'}return 'not found'})()"

Reading Results

After filling all three tabs, the app calculates:

Suspension output table:

  • Anti-Roll Bar (Front/Rear)
  • Damper Compression (Front/Rear)
  • Damper Extension (Front/Rear)
  • Spring Rate (Front/Rear)
  • Camber Angle (Front/Rear)
  • Toe Angle (Front/Rear)

Transmission output table:

  • Final Gear Ratio
  • Individual Gear Ratios (1st through 7th+)

Differential output table:

  • Initial Torque (Front/Rear)
  • Acceleration (Front/Rear)
  • Deceleration (Front/Rear)

Image Analyzer (Premium Feature)

GT Pro Tune has a built-in Image Analyzer that can extract dyno data from GT7 screenshots:

agent-browser click @e7  # Image Analyzer button
# Upload a screenshot of the dyno/power graph from GT7

Driving Style Presets

When the user describes a driving style, translate to suspension parameters:

StyleOversteer BiasLSDSpring StiffnessNotes
Aggressive / Verstappen+2 oversteer+2 LSD+1 stiffLate braking, throttle steering, rotate on entry
Balanced / Hamilton0 neutral00All-round consistency, predictable
Safe / Defensive-1 understeer-1 LSD-1 softStability first, forgiving on corner entry
Drift+3 oversteer+3 LSD-2 soft rearMaximum rotation, rear slides freely

Unit Conversions

  • Weight: 1 kg = 2.20462 lbs
  • Torque: 1 kgf·m = 7.233 ft-lb / 1 Nm = 0.7376 ft-lb
  • Power: 1 PS ≈ 0.9863 HP (close enough to use interchangeably for GT7)

GT7 Tuning Knowledge

See references/gt7-tuning-knowledge.md for:

  • Latest patch changes affecting tuning (1.55, 1.66)
  • Current meta (post-patch suspension, tire, LSD guidelines)
  • GT7-specific physics quirks vs real-world racing
  • Track-specific tuning workflows

Notes

  • All 581 GT7 cars are supported
  • Units: weight in lb, height in mm, speed in MPH, torque in ft-lb
  • The app calculates optimal settings — we input specs, it outputs the tune
  • GT7 Race Engineer (AI feature) requires a separate subscription tier
  • Always verify calculated tunes with a few laps in-game and adjust based on feel

Comments

Loading comments...