Who this is for, and the terms it keeps

Who can use it

• Drug discoveryA short list of experiments, each prepared as the recipe saidNatural-product labs usually screen a standard solvent extract. The Lab Planner picks a re-test panel from the 3 ranked hypotheses: plant–use pairs never tested the way they were made, with toxic species excluded. Artemisinin is the reference case: the activity was there once the preparation followed the old text.
• Journals and reviewersA five-item reporting checklistOf 43 studies that prepared the plant, 11 said whether heat was used and 7 whether the plant was fresh or dried. The checklist below costs authors a few lines of methods text and makes a negative result interpretable.
• Public healthA way to see what a home preparation containsPeople take plants as teas, decoctions, tinctures and oils, not as lab extracts. A study of one preparation can say little about another, and that cuts both ways: a compound of concern can be present in one and absent from the other. The Preparation Lab shows the predicted difference with its uncertainty range. It gives no guidance on preparing or taking any plant.
• Conservation of knowledge631 recipes kept with their exact wordsDatabases of traditional uses usually record a plant and a use, and drop the preparation. Here every recipe field is tied to a verbatim quote, so the how is kept alongside the what. A smaller, better-aimed set of experiments should also need less plant material to be collected for screening; we have not measured that.

A preparation reporting checklist (proposed)

The five details the audit codes, with what a methods section would state and how often the 43 studies that prepared the plant stated it. A sixth item applies whenever a study evaluates a traditional use: test at least one preparation made the traditional way, alongside the lab extract. Today, 6 of 257 article screenings (2.3%, 95% interval 1%–5%) compared two preparations of the same plant material.

Not independently validatedThe checklist is a proposal from this project; no journal has adopted it. Details: the evidence gap.

Traditional knowledge: whose knowledge, on what terms

• Only published, public-domain texts. The recipes come from Culpeper (1652 and 1850) and Pliny’s Natural History in a 19th-century translation. No knowledge was collected from a living community, so no community’s consent was needed and none was bypassed. This was a deliberate choice for a hackathon, where there is no time to build a partnership properly.
• These texts are themselves compilations. Both authors drew on earlier, often unnamed, practitioners. We credit the texts; we cannot credit those people, and we say so rather than treat the knowledge as authorless.
• Living traditions would need different terms. Before this method touches the knowledge of a living community, the project would follow the CARE Principles for Indigenous Data Governance (collective benefit, authority to control, responsibility, ethics), seek free, prior and informed consent through a community partnership, and agree access and benefit-sharing terms in line with the Nagoya Protocol. The community, not the project, would decide what is published. Our CC BY licence would not be the default for that data; Local Contexts Traditional Knowledge labels are one way to carry the community’s terms with it.
• No medical claims. The outputs are hypotheses worth re-testing. Every species in the study set carries a cited toxicity entry (not yet reviewed by a toxicologist), and plants with reported toxicity never get a Lab link, a plan action or a Planner slot.

Beyond the hackathon