These are real limits, not modesty. Practitioners raised every one of them, and they are accurate. A reference tool that will not say what it refuses to do is asking to be read as something it is not.

It does not differentiate patterns

Search constipation and AcuiQ will not ask whether the cause is heat, qi stagnation, cold, or deficiency – which is the question that determines point selection. It matches the word, not the presentation.

The index is not innocent of the distinction. It cannot act on it. A 2021 Chinese-language paper on point selection for constipation is stored here as six entries on one five-point base – ST25, BL25, ST37, TB06, KD06. The base alone is filed under constipation. The other five each add two points: LI04 and SP14 for heat, CV12 and LV03 for qi stagnation, CV04 and CV08 for cold, CV04 and BL20 for deficiency, CV04 and ST39 for dry stool. Each of those five is filed under its own term, such as constipation heat, and not under the bare complaint.

A search for constipation matches the word inside those terms as well, so pattern entries come back in the same list as the base entry, ranked by the words matched and the kind of source rather than by any pattern. Type the word into the search box and some of the pattern terms appear among the suggestions, ordered by how closely their names match what you typed. Nothing on either page says which of them describes you. Recording a differentiation and performing one are different acts, and only the first is in scope here.

Why that distinction is the whole of the discipline, and what it means for a symptom index, is set out in patterns, symptoms and what you can look up.

It does not ask what it would need to

Onset, duration, what makes the symptom better or worse, tongue, pulse, constitution – none of it is collected, so none of it informs a result. This is a deliberate consequence of the privacy position: no account, no login and no user record, which leaves nothing for an intake to be attached to. A search adds one row to a per-symptom tally, with a timestamp and no visitor on it, and that tally ranks the suggestions on the home page, not a result. A search that matches nothing records the term typed, with the day’s visitor code the privacy policy describes and nothing else about you, so the index can be extended. The privacy policy lists everything else the site records.

It does not explain why a point was chosen

The database records that a source paired a point with a symptom, not the reasoning behind the pairing. For that, use a point manual. Where a point’s own entry carries a caution, what that caution is worth is a separate question – reading a point warning covers it.

It is not a diagnosis and not a treatment plan

A published protocol worked for the patients in that study, under a diagnosis that study made. Neither transfers automatically to anyone else. An entry is a pointer to its citation; the judgement stays with a person.

Who it is for

Students and practitioners who want to see what the literature prescribes for a given complaint, and to get to the source quickly. It works well as a way into the literature and as a memory aid. It works badly as a substitute for differentiation, and it is not built for self-diagnosis by people without training.

Who maintains it

AcuiQ is maintained by AcuiQ. The index is compiled, corrected, and published in-house, and no entry on this site is written from clinical experience – every one of them records what a published source prescribed, which is why each carries its citation and why the limits above are stated as plainly as they are.

This is not clinical review

No practitioner signs off on individual entries. Extraction is checked against the source for attribution and for what the source claimed; it is not checked against clinical judgement, because the tool does not exercise any. Treat an entry as a pointer to its citation, not as a recommendation from AcuiQ.

Commercial interest

AcuiQ sells press needles, illustrated guides and point charts drawn from this same index, and a licence to the corpus as data. The site also has a per-request fee built for machines on three bulk exports of the corpus, paid over x402, a protocol that lets a program pay for a web request inside the request; it is switched off on the live site, which serves two of the exports free and does not serve the third, the full protocol file, at all.

Nothing about a result changes based on what is for sale: ranking depends on how closely an entry matches the words searched, the kind of source it came from and whether it names more than one point, as set out in how AcuiQ is built, and the guides and charts print protocols and points the site already shows you for free. But you are reading a reference tool published by a company that also sells needles, and you should weigh it accordingly.