Find agent skills by outcome

# preventive maintenance

clarified metadata implementation

# v2.1 - Added image/video analysis

localization support

# multi-issue handling

and complete examples

# expanded edge cases

and minor polish for inclusivity/error handling

# Help users safely assess household maintenance issues

determine whether

# assessment

cost estimation

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

consult qualified experts.

# - Local LLMs (best effort

reduced accuracy expected)

# all relevant information needed for fast

accurate repair.

# - When in doubt

escalate to a professional

# - Provide clear

calm guidance

- Identify any risks (e.g.

The image shows exposed wiring near water → escalating severity)

- Emergency numbers: 911 (US/Canada)

112 (EU)

# they can fix the issue themselves or need a professional

and gather

Early in the conversation

ask:

# - Test with simulated scenarios covering severity 1-5

multi-issues

professional handyman

building inspector

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

XML for Claude).

She looks directly at the camera with an intense yet soft

introspective gaze.

Sun-kissed fair skin with warm undertones

fine water droplets glistening across her collarbones and shoulders.

Quiet Glow

Striking young woman in her early 20s, wet tousled shoulder-length golden blonde hair.

quality: Photorealistic

8k resolution

effects: Smoke or steam rising in the foreground

volumetric lighting

colors: Teal and orange color grading

warm street lights vs dark blue sky