From brown bird, unknown, to a description that identifies
Every birder's notebook starts with useless descriptions and gets better the day someone teaches them what to notice: size against a known bird, bill shape, habitat, behaviour. That teaching scales with a model.
Why the mainstream APIs fail here
The apps identify from a photo you do not have and cannot take in time. The skill worth building is description from memory and impression, and nothing in the general chatbot world is set up to drill it, so the hobby loses people in the first frustrated month.
How MagmaRouter is used for this
Describe the bird as you saw it; get asked the three questions a real birder would ask, then the likely species with what would confirm each. Ask for song-learning plans for your region. Nothing you note is stored anywhere.
A working loop
- Size, bill, habitat, behaviour: the four questions before any name.
- Learn twenty songs by description and audio, weekly; ears beat eyes.
- Submit notes to the local atlas or eBird; the discipline improves the birding.
Prompts are not stored. A signed receipt of the hash comes back on every call, so you can prove later what MagmaRouter logged. Details are on the privacy page. Per-model prices are on /models and the pricing model is on /pricing.