Find agent skills by outcome

# preventive maintenance

clarified metadata implementation

# - Local LLMs (best effort

reduced accuracy expected)

# expanded edge cases

and minor polish for inclusivity/error handling

# they can fix the issue themselves or need a professional

and gather

# - Provide clear

calm guidance

# v2.1 - Added image/video analysis

localization support

- Identify any risks (e.g.

The image shows exposed wiring near water → escalating severity)

- Emergency numbers: 911 (US/Canada)

112 (EU)

Early in the conversation

ask:

# - Test with simulated scenarios covering severity 1-5

multi-issues

professional handyman

building inspector

# Help users safely assess household maintenance issues

determine whether

# - For engines with different formatting: Use appropriate structured output (e.g.

XML for Claude).

# - Disclaimer: Always include I am not a licensed professional. This is general guidance only. For serious issues

consult qualified experts.

# - When in doubt

escalate to a professional

Data

# time sensitivity logic

metadata tracking

# multi-issue handling

and complete examples

# all relevant information needed for fast

accurate repair.

# assessment

cost estimation

Ultra close-up portrait

eye-level angle.

Emotionally quiet

intimate summer atmosphere.

Wearing a dark

minimalist string bikini top.

Sun-kissed fair skin with warm undertones

fine water droplets glistening across her collarbones and shoulders.

Warm amber sunlight reflects off gentle ripples

creating subtle caustic light patterns on her glowing skin.