Tools
One tool that works, and thirteen designed but not built. Each of the thirteen says why — and in several cases the reason is that building it honestly needs data we do not have.
Available now
Conversions
Oven to air fryer converter
Converts an oven time and temperature to an air-fryer starting point, adjusted for your appliance architecture.
Cooking guidance
Cooking advisor
The long-term centre of the product: food, form, state, size and appliance in, a reasoned recommendation out.
Cost and energy
Running cost calculator
What a cook cost, given your own electricity rate — your metered figure if you have one, an upper bound from the rating plate if you do not.
Designed, not built
Writing down a tool’s inputs, outputs and limitations is most of the design work — and writing the limitations first is what stopped several of these being built this sprint.
- ConversionsTemperature converter
Celsius to Fahrenheit and back, with air-fryer-relevant reference points.
Why not yet: Not built because the conversion functions already exist in the domain layer and are used by the oven converter, which is the tool people actually arrive looking for. A standalone page for arithmetic anybody's phone will do is a page that exists to rank rather than to help, and building one in Sprint 1 would contradict the route contract.
- Cooking guidanceReheat advisor
Reheating guidance for a specific food, with the reheating safety rule applied.
Why not yet: Depends on the same observation catalogue as the main advisor and on a wider set of reheating records than Sprint 1 holds.
- Cost and energyAir fryer versus oven cost comparison
A side-by-side running-cost comparison for a specific cook.
Why not yet: Blocked on the same metered data as the running cost calculator.
- Capacity and sizingCapacity and portion estimator
How much food a given basket actually cooks well, in portions rather than litres.
Why not yet: Needs per-food single-layer area figures that do not exist yet. Establishing them is a measurement exercise rather than a research one, and it is a good candidate for the first controlled-testing sprint.
- Capacity and sizingBasket area estimator
Converts advertised litres into an estimated usable single-layer area.
Why not yet: Needs a body of measured capacity-to-area pairs across real models before the conversion is anything better than a guess. That comes from reading manuals and measuring baskets.
- Cooking guidanceDual-zone meal synchroniser
Works out when to start each drawer so two foods finish together.
Why not yet: Straightforward to build once the observation catalogue is broad enough that most pairs a reader might choose are actually covered. With nineteen observations it would refuse most requests.
- Safety referenceSafe temperature reference
The authoritative target for each food, with the published time–temperature equivalents.
Why not yet: The data model and the five seeded rules exist and each already has a page under /safety. A dedicated tool adds real value only once the rule catalogue covers materially more foods; until then it would be a search box over five records.
- Capacity and sizingAir fryer size selector
Recommends an architecture and size from how you actually cook.
Why not yet: Depends on the capacity estimator's underlying data. Also needs care to remain a guidance tool rather than becoming a product recommender, which would compromise the product's independence.
- CompatibilityMaterial compatibility checker
Search a material and get the conditional answer for your architecture.
Why not yet: The material pages already answer this well and each carries its per-architecture variations. A checker becomes worthwhile when the catalogue is large enough that browsing stops being practical.
- Safety referenceRecall checker
Check whether a specific model is subject to a recall or safety notice.
Why not yet: Requires a recall data feed AirfryerHQ does not have, and a commitment to keeping it current. Every model record in the catalogue currently reads `not_checked` for exactly this reason, which is the honest state. Building the tool before the feed exists would be the single most dangerous thing in this register.
- Cooking guidanceCooking time lookup
A fast search across every cooking observation in the catalogue.
Why not yet: Needs a search index and a broader observation catalogue. The cooking pages serve this need adequately at current scale.
Tools are not indexed. A tool’s output is assembled per query rather than written and reviewed, so there is no fixed page for a search engine to represent — and a safety-bearing answer nobody has read before it is shown is not something to submit to an index.