touri

AI Cost Tracking

🤖 LLM usage: $0.1666 (4 commits)
👤 Human dev: ~$201 (2.0h @ $100/h, 30min dedup)

Generated on 2026-06-15 using openrouter/qwen/qwen3-coder-next

touri maps new URI schemes/templates to reusable capabilities without creating a new library for every URI.

Core idea:

URI -> capability manifest -> backend -> service result

Example:

version: 1
capability:
  id: weather.forecast.html
  scheme: weather
  uri_template: weather://forecast/{place}/{days}/html
  operation: generate
  kind: command
backend:
  type: python
  target: python://examples.hello_weather:handler

Run:

pip install -e .
touri validate examples/20_touri_capabilities/weather_forecast.uri.capability.yaml
touri list examples/20_touri_capabilities
touri call weather://forecast/Gdansk/14/html --registry examples/20_touri_capabilities

Role in the system

URI -> capability manifest -> uri2run backend -> ServiceResult
touri does not replace uri3, uri2flow, uri2ops, or hypervisor.

Examples

Example	Path
Capabilities	`examples/20_touri_capabilities` (symlink → tellmesh)
Voice	`tellmesh/examples/21_touri_voice`

touri

AI Cost Tracking

Role in the system

Examples

Links