# askWheat (remote · askwheat.plantdatabases.org)

Wheat genes, Watkins pangenome, EMS mutants, expression, genotypes, germplasm and papers (WheatIS)

- Trust score: 66/100 (medium)
- Change this week: +3
- Registry status: active
- Liveness: live
- Owner verified: no
- Last scored: 2026-11-21

## Components

- remote · `askwheat.plantdatabases.org`: 66/100 (this document), [markdown](https://verifymcp.io/servers/org-plantdatabases-askwheat/askwheat.md), [page](https://verifymcp.io/servers/org-plantdatabases-askwheat/askwheat)

## Channel facts

- Endpoint: `https://askwheat.plantdatabases.org/mcp`
- Transports: `streamable-http`
- Auth: `none`
- Version: `1.0.0`

## Trust breakdown

How this component scores in each security and reliability category. Every signal is checked automatically against the live server, and we only credit what we can confirm. Scores are 0–100 per category. Scoring method: https://verifymcp.io/docs/scoring (what has changed: https://verifymcp.io/docs/scoring/changelog)

Scored 2026-11-21.

- **Endpoint Security**: 80/100
  - The endpoint's TLS certificate is valid, in date, and uses a strong key.
  - No authorisation is required to call this server. Nothing on it is marked destructive, but we can't show that an anonymous caller can only read: full credit needs the server to list at least one tool and every tool to be verifiably read-only. We use a best-effort check of tool names and descriptions to tell read-only tools from tools that change things, so some genuinely read-only tools may not get full credit. Names that start with a read verb (get, list, read, search and similar) and accurate readOnlyHint annotations help. Meet that to earn full credit, or require authorisation.
  - HTTPS is enforced; there's no plaintext access path.
  - The HSTS (Strict-Transport-Security) header is present.
  - DNSSEC check failed: this domain isn't protected by DNSSEC.
- **Transport & Reachability**: 100/100
  - Verified streamable-http transport via a live MCP handshake.
- **Schema Quality & AI Usability**: 32/100
  - 0% of prompts and resources have a non-trivial description (not blank, and not just the item's name).
  - AI-judged instruction clarity (good).
  - Context-footprint check failed: tool/resource definitions use about 15475 tokens (~230/item across 67 items; 65 tools + 2 resources), over budget; trim descriptions and params.
  - Usage-examples check failed: none of the tools include examples.
- **Stability & Change Management**: 23/100
  - Stability observed for 7 of 30 days with no destabilising changes; credit accrues until the full window elapses.
- **Tool Coverage**: 67/100
  - 100% of tools have a non-trivial description (not blank, and not just the tool's name).
  - 0% of tool parameters carry a description.
- **Tool Safety**: 100/100
  - No prompt-injection markers were found in the server instructions, tool names or descriptions we captured.
  - We read all 65 captured tool definition(s), and no name or description among them implies an irreversible operation.
  - An AI judge read all 67 captured unit(s) of tool text and found none that tries to manipulate the model reading it.
- **Capabilities**: 100/100
  - Implements a current MCP spec version (2026-07-28).
  - Supports UI / widget rendering.

## Install

### How do I install the askWheat MCP server?

askWheat is a hosted endpoint at https://askwheat.plantdatabases.org/mcp, so there is nothing to install locally. Ready-made configuration for Claude, Cursor, VS Code, Codex and 5 more is on this page, copied from each client's own documentation.

### Claude

```bash
claude mcp add --transport http org-plantdatabases-askwheat 'https://askwheat.plantdatabases.org/mcp'
```

### Cursor

```json
{
  "mcpServers": {
    "org-plantdatabases-askwheat": {
      "url": "https://askwheat.plantdatabases.org/mcp"
    }
  }
}
```

### VS Code

```json
{
  "servers": {
    "org-plantdatabases-askwheat": {
      "type": "http",
      "url": "https://askwheat.plantdatabases.org/mcp"
    }
  }
}
```

### Codex

```toml
[mcp_servers.org-plantdatabases-askwheat]
url = "https://askwheat.plantdatabases.org/mcp"
```

### opencode

```json
{
  "$schema": "https://opencode.ai/config.json",
  "mcp": {
    "org-plantdatabases-askwheat": {
      "type": "remote",
      "url": "https://askwheat.plantdatabases.org/mcp",
      "enabled": true
    }
  }
}
```

### OpenClaw

```bash
openclaw mcp add org-plantdatabases-askwheat --url 'https://askwheat.plantdatabases.org/mcp' --transport streamable-http
```

### Hermes

```yaml
mcp_servers:
  org-plantdatabases-askwheat:
    url: "https://askwheat.plantdatabases.org/mcp"
```

### Netclaw

```json
{
  "McpServers": {
    "org-plantdatabases-askwheat": {
      "Transport": "http",
      "Url": "https://askwheat.plantdatabases.org/mcp"
    }
  }
}
```

### Vellum

```bash
assistant mcp add org-plantdatabases-askwheat -t streamable-http -u 'https://askwheat.plantdatabases.org/mcp'
```

### Other

```json
{
  "mcpServers": {
    "org-plantdatabases-askwheat": {
      "type": "http",
      "url": "https://askwheat.plantdatabases.org/mcp"
    }
  }
}
```

The mcpServers block is a cross-client convention. Remote transports vary, so check your client's docs.

## Changelog

Every change recorded for this component, newest first. Days that predate change tracking, or that we cannot explain, say so: "we were watching and nothing happened" and "we were not watching" are different claims.

### 2026-11-21 (score 66, 0)

- [security] The server rewrote its instructions, which are the text every model session reads
- [functional] New tool “editing_candidates”
- [functional] New tool “gwas_associations”

### 2026-11-20 (score 66, 0)

- [security] The server rewrote its instructions, which are the text every model session reads
- [functional] New tool “cereal_pangenome_presence”
- [functional] New tool “cereal_trait_genes”
- [functional] New tool “kasp_assays”

### 2026-11-19 (score 66, +1)

- [security] The server rewrote its instructions, which are the text every model session reads
- [functional] New tool “wheat_segment_pav”

### 2026-11-18 (score 65, 0)

- [security] The server rewrote its instructions, which are the text every model session reads
- [security] Tool “brapi_germplasm_studies” rewrote its description, which is the text the model reads
- [security] Tool “brapi_observations” rewrote its description, which is the text the model reads
- [functional regression] Schema quality: 12233 → 13624
- [functional] New tool “genesys_accession”
- [functional] New tool “watkins_extra_genes”
- [functional] New tool “watkins_nlr_cluster”
- [functional] New tool “wheat_gene_in_cultivars”

### 2026-11-17 (score 65, +1)

- [security] The server rewrote its instructions, which are the text every model session reads
- [security] Tool “data_sources” rewrote its description, which is the text the model reads
- [security] Tool “gene_expression” rewrote its description, which is the text the model reads
- [security] Tool “trait_hits” rewrote its description, which is the text the model reads
- [functional] New tool “rice_function”
- [cosmetic] “gene_expression” added an optional parameter “max_experiments”

### 2026-11-15 (score 64, +1)

- [security] The server rewrote its instructions, which are the text every model session reads
- [security] Tool “trait_hits” rewrote its description, which is the text the model reads
- [functional regression] Schema quality: 10776 → 11892
- [functional improvement] Stability: unverified → 0.03
- [functional] New tool “annotate_genes”
- [functional] New tool “blast_sequence”
- [functional] New tool “gene_neighbourhood”
- [functional] New tool “gene_papers”
- [functional] New tool “genes_with_annotation”

### 2026-11-14 (score 63)

First indexed and scored.

## MCP tools (65)

### `search_data` (~146 tokens)

Full-text search across all askWheat resources (genes, germplasm, genotyping datasets, files).

Filters (exact values): database e.g. 'Ensembl Plants', 'Watkins Wheat Pangenome', 'CropSNPdb', 'GBIF genebank accessions';
entry_type e.g. 'Gene annotation', 'Germplasm', 'Genotyping Experiment', 'File';
species e.g. 'Triticum aestivum'. Returns total count, facet counts and the top hits with source URLs.

Input parameters:

- `database`
- `entry_type`
- `limit` (integer)
- `query` (string, required)
- `species`

### `find_germplasm` (~186 tokens)

Find wheat accessions/cultivars/lines by name or accession ID across five indexed sources: CropSNPdb
(SNP-genotyped lines, incl. all Australian Grains Genebank wheat and CAIGE breeding lines), the Watkins Wheat
Pangenome (sequenced landraces), OzWheat (CSIRO GWAS panel lines), FAIDARE (INRAE GnpIS collections, e.g. the
FRA040 Clermont-Ferrand wheat collection, and EVA European genebank evaluation sets) and genebank accessions of
Triticum and Aegilops from GBIF (holding genebank as FAO WIEWS code, e.g. MEX002 = CIMMYT, accession number;
mostly Svalbard safety duplicates).

Input parameters:

- `limit` (integer)
- `name` (string, required)

### `data_sources` (~58 tokens)

What askWheat currently indexes: record counts per database and data type, and data versions; plus the live
endpoints it queries or offers (BrAPI nodes, the CropSNPdb genotype API, its own MCP server and JSON API).

### `resolve_gene` (~110 tokens)

Resolve a wheat gene symbol (e.g. Lr34, Rht-B1) or IWGSC gene ID in any RefSeq version
(v1.0 TraesCS..01G, v1.1 TraesCS..02G, v2.1 TraesCS..03G, CSS2014 Traes_..) to all versions,
genomic locations, curated symbol/trait (with verification status), Watkins pangenome presence and links.

Input parameters:

- `identifier` (string, required)

### `gene_function` (~76 tokens)

IWGSC RefSeq v2.1 functional annotation (InterPro, Pfam, Gene Ontology) for a gene symbol or any-version
IWGSC ID. annotation_types filters e.g. ['Gene Ontology'] or ['Pfam','InterPro'].

Input parameters:

- `annotation_types`
- `identifier` (string, required)

### `genes_in_region` (~107 tokens)

List IWGSC genes overlapping a region. chromosome like '7D'; assembly 'v2.1' (RefSeq v2.1 coordinates)
or 'v1.0' (RefSeq v1.0 coordinates, v1.1 gene IDs). Region size is capped at 20 Mb.

Input parameters:

- `assembly` (string)
- `chromosome` (string, required)
- `end` (integer, required)
- `limit` (integer)
- `start` (integer, required)

### `find_marker` (~131 tokens)

Position(s) of a genetic marker (90K/80K/9K iSelect, 35K Axiom, KASP, DArT, SSR e.g. 'BS00000713', 'wmc617', 'Xgwm149';
40K array IDs such as 'IWB61980' or 'AVRIG14047' come from CropSNPdb)
on IWGSC RefSeq v2.1 (IWGSC GMAP alignments), with genes within +/- flank bp of each hit.

Input parameters:

- `flank` (integer)
- `marker` (string, required)

### `wheat_homologues` (~217 tokens)

Homoeologues (the A/B/D copies), and orthologues in other grasses and Arabidopsis, of a bread wheat gene from
Ensembl Plants Compara gene trees. gene: symbol or any IWGSC ID (converted to RefSeq v1.1). By default
orthologues are listed for durum and wild emmer, T. urartu, Ae. tauschii, rye, barley, oat, Brachypodium, rice,
maize, sorghum and Arabidopsis; target_species takes Ensembl names instead (e.g. ['triticum_aestivum_mace',
'hordeum_vulgare_goldenpromise'] for pan-genome cultivars). include_paralogues adds within-wheat paralogues
(same-genome duplicates). Relationships: homoeolog_one2one, ortholog_one2one/one2many/many2many.

Input parameters:

- `gene` (string, required)
- `include_paralogues` (boolean)
- `target_species`

### `wheat_gene_in_cultivars` (~224 tokens)

The corresponding genes of a Chinese Spring gene in the other bread wheat genome assemblies of the 21-genome
pangenome: Ensembl Plants Compara orthologues in ArinaLrFor, Jagger, Julius, Kariega, Lancer, Landmark, Mace,
Mattis, Norin 61, Paragon, Renan and Stanley, each with its gene ID, chromosome position and a link to this
site's genome browser for that assembly. gene: symbol or any IWGSC ID. same_subgenome_only keeps the copy on the
same subgenome as the query (Compara also lists the homoeologous copies). Cultivars without a Compara orthologue,
and assemblies outside Compara (Cadenza, Claire, Robigus, Weebil, Fielder, Zang1817, Attraktion, Aikang 58), are
listed: search those with blast_sequence against their genome database.

Input parameters:

- `gene` (string, required)
- `same_subgenome_only` (boolean)

### `rice_function` (~182 tokens)

Function of a wheat gene through its rice orthologue(s): Ensembl Plants Compara orthologues in Oryza sativa
Japonica (RAP IDs, Os03g0707600), then Gramene's curated rice gene name and synonyms (e.g. SLR1, Hd1), Trait
Ontology terms (phenotypes linked to the gene), QTL traits, Plant Ontology (where/when studied), Plant Reactome
reactions and PubMed IDs. gene: a wheat symbol or any IWGSC ID, or a rice RAP ID directly. Rice is the best
studied cereal, so this often names a wheat gene that has no curated function of its own; the wheat gene's
function is inferred, not shown. All one2many / many2many orthologues are listed.

Input parameters:

- `gene` (string, required)

### `wheat_mutants` (~306 tokens)

Induced mutations and natural variants in and around a bread wheat gene. source 'ems' (default): EMS mutant
lines of tetraploid Kronos and hexaploid Cadenza with exome-capture sequencing (Krasileva et al. 2017; 2,735
lines), giving the line ID to order, e.g. Cadenza1158; 'diversity': exome-capture variants of 890 landraces and
cultivars (He et al. 2019) and 103 Watkins landraces; 'all'. Every variant Ensembl assigns to the gene is kept:
coding, splice, UTR, intron and upstream/downstream (within 5 kb). consequence filters: 'truncating' (stop gained,
splice donor/acceptor, start/stop lost), 'missense', 'coding', 'splice', 'utr', 'intron', 'flanking'
(upstream/downstream), 'regulatory' (flanking + UTR) or a Sequence Ontology term. homoeologues=True adds the A/B/D
copies (to stack knockouts). gene: symbol or any IWGSC ID (converted to RefSeq v1.1); positions are RefSeq v1.0.

Input parameters:

- `consequence`
- `gene` (string, required)
- `homoeologues` (boolean)
- `limit` (integer)
- `source` (string)

### `wheat_variants_in_region` (~160 tokens)

EMS mutations and exome-capture variants in a bread wheat interval on IWGSC RefSeq v1.0, including intergenic
ones (promoters, unannotated regions, off-target capture). region: 'chr:start-end', e.g. '3A:100000000-100500000'
(max 5 Mb). source 'ems', 'diversity' or 'all' (default); consequence as in wheat_mutants, plus 'intergenic'.
Genic variants name the gene (RefSeq v1.1); use genes_in_region for the genes near intergenic ones.

Input parameters:

- `consequence`
- `limit` (integer)
- `region` (string, required)
- `source` (string)

### `wheat_mutant_line` (~176 tokens)

Every mutation Ensembl lists for one EMS line (Cadenza or Kronos exome-capture TILLING; Krasileva et al. 2017),
e.g. 'Cadenza1166' or 'Kronos2235': counts by consequence, the genes with truncating mutations, and the mutations
(gene, consequence, amino-acid change, RefSeq v1.0 position), most severe first. consequence filters as in
wheat_mutants ('truncating', 'missense', ...); chromosome e.g. '6D'. Use it to judge a line's background load and
to check for off-target hits near a gene of interest before crossing.

Input parameters:

- `chromosome`
- `consequence`
- `limit` (integer)
- `line` (string, required)

### `wheat_knockout_screen` (~217 tokens)

Which genes have EMS alleles in every homoeologue, i.e. can be knocked out across A, B and D by crossing
Cadenza/Kronos lines. For each gene (genes: up to 300 symbols or IWGSC IDs, or region: RefSeq v1.0 interval up to
10 Mb, e.g. a QTL) the A/B/D copies come from Ensembl Compara, and each copy gets its lines with a consequence
'truncating' (default; stop, splice donor/acceptor, start/stop lost) or 'missense' etc. Status 'all_copies',
'partial' or 'none'; complete_only keeps 'all_copies'. Kronos lines cover only A and B copies; D copies need
Cadenza. Lines hitting two copies at once are flagged.

Input parameters:

- `complete_only` (boolean)
- `consequence` (string)
- `genes`
- `limit` (integer)
- `region`

### `watkins_gene_presence` (~90 tokens)

Presence/absence of one gene across the 1,038 Watkins pangenome accessions: present count, core/variable,
counts by ancestral group (AG1-AG7, Modern), continent and growth habit, UniProt/GO annotation and links.
gene: symbol, any IWGSC RefSeq ID, or a pangenome gene-model ID.

Input parameters:

- `gene` (string, required)

### `watkins_gene_accessions` (~115 tokens)

Watkins accessions that carry (present=True) or lack (present=False) a gene, optionally filtered by
country, continent, ancestral_group (AG1-AG7 or Modern) or growth_habit. Returns metadata, coordinates and
Seedstor ordering links.

Input parameters:

- `ancestral_group`
- `continent`
- `country`
- `gene` (string, required)
- `growth_habit`
- `limit` (integer)
- `offset` (integer)
- `present` (boolean)

### `watkins_compare_groups` (~133 tokens)

Genes whose presence frequency differs most between two sets of Watkins accessions.
group_a / group_b as '<field>:<value>' with field one of group (ancestral group), country, continent,
growth_habit, e.g. 'group:AG3' vs 'group:Modern', 'country:India' vs 'country:China'.
min_diff is the minimum absolute frequency difference (0-1). Negative diff = more common in group_b.

Input parameters:

- `group_a` (string, required)
- `group_b` (string, required)
- `limit` (integer)
- `min_diff` (number)

### `watkins_find_genes` (~75 tokens)

Find Watkins pangenome genes by UniProt protein name, gene name, protein family or ID fragment
(e.g. 'FAR1', 'NLR', 'Disease resistance protein RGA'). Returns gene IDs with UniProt/GO annotation.

Input parameters:

- `limit` (integer)
- `query` (string, required)

### `watkins_accession` (~98 tokens)

One Watkins accession (e.g. WATDE0063, or a synonym): origin, ancestral group, coordinates, Seedstor link,
number of genes present, presence/absence of the listed genes (symbols or any IWGSC IDs accepted), all its
phenotype records (trial, trait, value, units) and any RIL populations it is a parent of.

Input parameters:

- `genes`
- `name` (string, required)

### `watkins_phenotypes` (~147 tokens)

Field/glasshouse phenotypes of Watkins accessions (heading, height, grain size/weight/quality, senescence,
lodging, rust and mildew scores, morphology). With no arguments, lists the trials (JIC UK 2006-2020, Egypt
2019-2021, Limagrain seedling yellow rust 2020, Kenya/Ethiopia rust 2013) and trait names with units.
Filter by accession (e.g. WATDE0063), exact trait name, trial and/or country.

Input parameters:

- `accession`
- `country`
- `limit` (integer)
- `trait`
- `trial`

### `watkins_gene_trait` (~111 tokens)

Compare a trait between Watkins accessions that carry vs lack a gene, per trial: n, mean and median per
group and a Mann-Whitney p-value (exploratory; no correction for population structure). gene: symbol, any
IWGSC RefSeq ID or pangenome gene-model ID. trait: exact name from watkins_phenotypes (e.g. 'Heading time').

Input parameters:

- `gene` (string, required)
- `trait` (string, required)
- `trial`

### `watkins_populations` (~112 tokens)

Paragon x Watkins (and a few Robigus/other x Watkins) RIL populations: Watkins parent, country, ancestral
group, number of RILs, GRU seed code, genetic map type and whether it donated TK-NILs. With population (e.g.
'ParW034') or watkins_parent (e.g. 'WATDE0005') also lists the trials in which its lines were phenotyped.

Input parameters:

- `population`
- `watkins_parent`

### `watkins_population_phenotypes` (~166 tokens)

Trial phenotypes (line means per trial) of Paragon x Watkins RILs and TK-NILs: heading, height, yield, grain
traits, biomass, N content and more from JIC, Rothamsted and Nottingham trials (2011-2022). population_type 'RIL',
'NIL' or 'check' (parents and controls, e.g. Paragon). line matches by prefix (e.g. 'ParW034_0012' also returns NILs
derived from it). With no arguments, lists trials and traits.

Input parameters:

- `limit` (integer)
- `line`
- `population`
- `population_type`
- `trait`
- `trial`
- `watkins_parent`

### `watkins_trait_genes` (~188 tokens)

Which variable genes differ in presence between Watkins accessions with high vs low values of a trait in one
trial (e.g. late vs early heading). Accessions are split into top/bottom quartile (split: quartile, tercile,
median) within each ancestral group (stratify_by_group) and every variable gene is tested (Cochran-Mantel-Haenszel,
Benjamini-Hochberg FDR). direction 'high' = genes more frequent in high-trait accessions, 'low' the reverse.
trait and trial are exact names from watkins_phenotypes(). Exploratory screen, not a GWAS.

Input parameters:

- `direction` (string)
- `limit` (integer)
- `split` (string)
- `stratify_by_group` (boolean)
- `trait` (string, required)
- `trial` (string, required)

### `watkins_nil_genes` (~159 tokens)

Genes inside the Watkins introgression of a TK-NIL (Watkins segment in a Paragon background), with presence in
the Watkins donor and frequency in modern cultivars and landraces. nil_line is a NIL line name from
watkins_population_phenotypes (e.g. 'ParW141_0016-7-10-Q2D-Hd-P'). The interval comes from the NIL name
(chromosome-start-end on RefSeq v1.0). Highlights genes the donor lacks but most modern cultivars carry, and
genes the donor carries but most modern cultivars lack (Paragon itself is not in the pangenome).

Input parameters:

- `limit` (integer)
- `nil_line` (string, required)

### `watkins_extra_genes` (~363 tokens)

Genes found on the Watkins pangenome extra contigs: sequence not in the 21 wheat genome assemblies, assembled
from the reads of 827 Watkins landraces and 217 modern cultivars (163,778 gene models; 34,295 with a confident
UniProt match, of which 10,426 occur only in landraces). Use it for genes that modern wheat lacks.
gene: one gene model ID (e.g. maker-asm3_jcf7180084687145-snap-gene-0.8) -> its contig, annotation, presence by
ancestral group and the accessions carrying it. Otherwise lists genes: query = words or a UniProt accession matched
against gene name, protein name and family (e.g. 'NLR', 'disease resistance protein RPM1', 'Q9SZL8');
presence = landrace_only | modern_only | shared | any; group = ancestral group the gene must be present in (AG1-AG7,
Modern); nlr = only genes overlapping the 176 landrace-specific NLR loci (NLR-Annotator); disease / stress = GO-based
flags; annotated_only = only genes with a UniProt match; min_accessions = minimum number of carriers.
Absence means reads did not cover the gene model (divergence or low coverage are possible); rare genes were
assembled from few reads.

Input parameters:

- `annotated_only` (boolean)
- `disease`
- `gene`
- `group`
- `limit` (integer)
- `min_accessions` (integer)
- `nlr` (boolean)
- `presence` (string)
- `query`
- `stress`

### `watkins_nlr_cluster` (~133 tokens)

A landrace-specific NLR cluster in the Watkins pangenome: nine NLR loci that co-occur in 58
Watkins landraces (mainly AG4, AG3, AG6; Afghanistan, former USSR, Greece, Iran, Spain) and are absent from all 217
modern cultivars, including a complete CC-NBARC-LRR gene (best UniProt match rice blast resistance protein Pikm2),
placed by read-pair links near TraesCS2A02G590100 on distal 2AL. Returns the 58 carriers and the NLR genes.

### `wheat_segment_pav` (~356 tokens)

Which of the 21 wheat genome assemblies of the UWA graph pangenome (Chinese Spring, ArinaLrFor, Cadenza, Claire,
Jagger, Julius, Kariega, Lancer, Landmark, Mace, Mattis, Norin61, Paragon, Renan, Robigus, Stanley, Weebil, Fielder,
Zang1817, Attraktion, Aikang58) carry each stretch of graph sequence: presence/absence variation beyond genes.
Region mode: region = 'chrom:start-end' on Chinese Spring RefSeq v1.0 (<= 10 Mb), or gene = a symbol / IWGSC ID
(its v1.0 position +- flank). Returns how much Chinese Spring sequence each assembly lacks there, the variable
intervals (largest first), and non-reference sequence (present in other assemblies, not Chinese Spring) inserted
in the region, anchored by the graph's flanking Chinese Spring segments. assembly = another assembly id (e.g.
Wheat_Cadenza) with region on that assembly's own sequence names: only the non-reference graph sequence that came
from that assembly has coordinates there.
Group mode: present_in / absent_in = cultivar lists -> intervals in all of present_in and none of absent_in
(genome-wide, or one chromosome), with totals and the largest intervals. Can combine with region or gene.
min_length = smallest interval reported (bp).

Input parameters:

- `absent_in`
- `assembly`
- `chromosome`
- `flank` (integer)
- `gene`
- `limit` (integer)
- `min_length` (integer)
- `present_in`
- `region`

### `watkins_gene_haplotypes` (~223 tokens)

Haplotypes of a gene across the Watkins landraces and modern cultivars (WatSeq, 1,051 accessions). 'snp'
(default where built): accessions grouped by their SNP alleles in the gene +-2 kb (MAF >= 0.02), each haplotype
with its coding changes (e.g. p.Arg134Gln), whether it is Chinese Spring-like, size and frequency by ancestral
group (AG1-AG7, Modern) and country; 'ibspy': the IBSpy haplotype group of the 1 Mb block holding the gene, with
the pangenome assemblies (e.g. jagger-pg, mace-pg) that share each block haplotype. Accessions lacking the gene in
the Watkins pangenome are listed as 'absent'. gene: symbol or any IWGSC ID (positions RefSeq v1.0).

Input parameters:

- `gene` (string, required)
- `members` (integer)
- `source` (string)

### `watkins_haplotype_trait` (~172 tokens)

Trait values per haplotype of a gene (watkins_gene_haplotypes) in each Watkins trial: n, mean and median per
haplotype and a Kruskal-Wallis test across haplotypes with >= min_n accessions. Exploratory: no correction for
population structure or multiple testing, so a difference can reflect ancestry rather than the gene. trait: exact
name from watkins_phenotypes (e.g. 'Heading time'); trial optional (all trials otherwise). include_values adds
each accession's mean per haplotype (for plotting).

Input parameters:

- `gene` (string, required)
- `include_values` (boolean)
- `min_n` (integer)
- `source` (string)
- `trait` (string, required)
- `trial`

### `watkins_lof_alleles` (~135 tokens)

Watkins landraces and modern cultivars carrying a predicted loss-of-function allele of a gene (WatSeq, 1,051
accessions): each stop-gained, start/stop-lost or splice-site SNP with its protein change, the number of
homozygous and heterozygous carriers, and the carriers (accession, ancestral group, country) for ordering from the
JIC Germplasm Resources Unit. gene: symbol or any IWGSC ID; homoeologues are listed for checking redundancy.

Input parameters:

- `carriers` (integer)
- `gene` (string, required)

### `watkins_accession_lof` (~139 tokens)

Genes with a predicted loss-of-function allele (stop gained, start/stop lost, splice site) in one Watkins
landrace or modern cultivar (WatSeq). accession: Watkins store code or WatSeq sample name (e.g. 'Watkins 1190027',
'WATDE0027', 'Paragon'). genes optionally restricts to a gene list (symbols or IWGSC IDs). Rare alleles (carried by
few accessions) are listed first.

Input parameters:

- `accession` (string, required)
- `genes`
- `homozygous_only` (boolean)
- `limit` (integer)

### `expression_experiments` (~114 tokens)

Bread wheat RNA-seq experiments held by askWheat (Expression Atlas and SRA study SRP045409 (seedling drought, heat and combined stress; quantified by askWheat)): baseline (TPM by
tissue / developmental stage / treatment) and differential (stress, pathogen, genotype contrasts). query filters descriptions/labels (e.g. 'drought',
'grain', 'Fusarium'). Gives each experiment's assay groups (baseline) or contrasts (differential).

Input parameters:

- `query`

### `gene_expression` (~190 tokens)

Where and when a wheat gene is expressed, and under which treatments it changes, from Expression Atlas and SRA study SRP045409 (seedling drought, heat and combined stress; quantified by askWheat).
gene: symbol or any IWGSC RefSeq ID (converted to v1.1, as used by Atlas). Baseline: TPM (median over replicates,
with range) for the top_groups highest-expressed tissues/stages per experiment. Differential: every contrast where
the gene is significant (adjusted p < 0.05, |log2FC| >= 1). experiment restricts to one accession; otherwise the
max_experiments baseline experiments with the highest TPM are detailed and the rest listed with their maximum.

Input parameters:

- `experiment`
- `gene` (string, required)
- `max_experiments` (integer)
- `top_groups` (integer)

### `expression_genes` (~170 tokens)

Genes characteristic of a tissue/stage or responsive to a treatment in one Expression Atlas experiment.
Baseline experiment + group (text matched to group labels, e.g. 'root', 'grain'): genes with TPM >= min_tpm in a
matching group ranked by specificity = (max TPM in matching groups + 1) / (max TPM in other groups + 1).
Differential experiment + contrast (text matched to contrast names or IDs): significant genes, direction 'up' or
'down', ranked by |log2FC|. Use expression_experiments() for group and contrast names.

Input parameters:

- `contrast`
- `direction` (string)
- `experiment` (string, required)
- `group`
- `limit` (integer)
- `min_tpm` (number)

### `wheat_homoeologue_expression` (~125 tokens)

Relative expression of a wheat gene's A, B and D homoeologues per tissue/stage (Expression Atlas baseline
experiments), with the homoeologue expression balance category of each tissue: balanced, A/B/D dominant or
A/B/D suppressed, by distance to the ideal proportions (Ramirez-Gonzalez et al. 2018). Categories are given only
for 1:1:1 triads with summed TPM >= 0.5. experiment restricts to one accession.

Input parameters:

- `experiment`
- `gene` (string, required)

### `snp_lines` (~117 tokens)

Find SNP-genotyped lines in CropSNPdb by name or AGG accession (e.g. 'Cadenza', 'AGG 38862', 'CAIGE ZWB17' or a CAIGE GID '7396745').
Returns line name, accession(s), country, pedigree, status, breeding organisation and donor. dataset: 'wheat40k' (default) or 'wheat90k'.

Input parameters:

- `dataset` (string)
- `limit` (integer)
- `query` (string, required)

### `snp_genotypes` (~88 tokens)

SNP genotype calls for up to 50 lines in a region (chromosome like '2D'). Returns markers with positions
and each line's call. Keep regions small (a few Mb) for many lines.

Input parameters:

- `chromosome` (string, required)
- `dataset` (string)
- `end` (integer)
- `lines` (array, required)
- `start` (integer)

### `snp_compare_lines` (~91 tokens)

Compare two lines' SNP genotypes on a chromosome (optionally a region): markers compared, identical,
different, % identity and the regions where they differ (useful for pedigree/introgression questions).

Input parameters:

- `chromosome` (string, required)
- `dataset` (string)
- `end`
- `line_a` (string, required)
- `line_b` (string, required)
- `start`

### `snp_haplotype_carriers` (~190 tokens)

Lines that share a reference line's SNP haplotype in a region (max 5 Mb), e.g. which genebank lines carry the
Cadenza haplotype around a gene. Only markers polymorphic in the panel (minor allele frequency >= min_maf) are
used, and a match needs calls at >= min_called of them with at most max_mismatch differences. Always report
informative_markers and fraction_matching; if 'note' says the haplotype is common, say it is not diagnostic.

Input parameters:

- `chromosome` (string, required)
- `dataset` (string)
- `end` (integer, required)
- `limit` (integer)
- `line` (string, required)
- `max_mismatch` (integer)
- `min_called` (number)
- `min_maf` (number)
- `start` (integer, required)

### `snp_markers` (~62 tokens)

SNP array markers (IDs and positions) in a region, with the total count.

Input parameters:

- `chromosome` (string, required)
- `dataset` (string)
- `end` (integer, required)
- `limit` (integer)
- `start` (integer, required)

### `snp_phenotypes` (~121 tokens)

Field trial trait data held in CropSNPdb for this crop (T3 trials with Crop Ontology trait IDs, Australian Grains
Genebank records). With no arguments, lists the available traits. Filter by line and/or trait: trait matches the
exact name or any part of it (e.g. 'Height', 'grain yield'). Most phenotyped lines are in the
wheat90k dataset.

Input parameters:

- `dataset` (string)
- `limit` (integer)
- `line`
- `trait`

### `ozwheat_mtas` (~233 tokens)

Marker-trait associations from the OzWheat GWAS (285 Australian and international wheat cultivars and lines;
glasshouse 2016 and field trials at Canberra 2018-19, Wagga 2018, Narrabri 2019). Traits: days to anthesis,
days to heading, degree-days to heading ('flowering'/'heading') and plant height. Filter by trait, gene (symbol or
any IWGSC ID; matches the gene carrying the SNP or the linked candidate gene, or the name in their annotation) and/or region (chromosome like '7B'
with start/end in RefSeq v1.0 bp). Each MTA gives the peak SNP and its position, p-value, the gene it sits in
(with v1.1/v2.1 IDs), any literature candidate gene, and the allele effect: trait mean (BLUE) of lines carrying
each allele in that trial.

Input parameters:

- `chromosome`
- `end`
- `gene`
- `limit` (integer)
- `start`
- `trait`

### `ozwheat_lines` (~194 tokens)

OzWheat diversity panel lines (285 Australian and international cultivars and breeding lines): name, OzWheat ID,
Australian Grains Genebank (AGG) accession, pedigree and phenotype BLUEs (days to anthesis/heading, degree-days to
heading, height per trial). query matches name, OzWheat ID (oz0001), AGG number or pedigree words. marker (an
OzWheat MTA peak SNP such as 'SNP34815_7B', from ozwheat_mtas) with allele lists the lines carrying that allele.
include_genotypes adds each line's calls at all MTA peak markers. Lines are also often in CropSNPdb under the
same cultivar name (use snp_lines).

Input parameters:

- `allele`
- `include_genotypes` (boolean)
- `limit` (integer)
- `marker`
- `query`

### `genebank_phenotypes` (~369 tokens)

Field phenotypes of wheat cultivars and genebank accessions from IPK e!DAL (CC BY 4.0). Datasets: 'GABI-WHEAT'
(372 European elite cultivars 1975-2007; 16 traits incl. grain yield, heading, height, TKW, protein, sedimentation,
falling number, FHB, Septoria, tan spot; BLUEs per trial environment and across 2009-2010 trials in Germany/France),
'IPK' (12,754 IPK genebank accessions, TRI numbers: flowering time, plant height, TKW, BLUEs across 70 years of
regeneration), 'EU-GENEBANKS' (about 40,000 accessions of CREA, ICARDA, IHAR, INIA, IPGR, NARDI, NPPC, Agroscope:
heading, height, TKW, some disease, protein and yield). No arguments lists datasets and traits.
query: a cultivar name or accession number (e.g. 'Julius', 'TRI 1000', 'IG 107099') -> all its trait values.
trait (+ order 'high'/'low'): rank accessions of one dataset. filters: several traits at once, e.g.
'plant height < 90; thousand kernel weight > 45' (BLUEs, one dataset). genebank: FAO code or name (LBN002, CREA);
country: origin (IPK dataset).

Input parameters:

- `country`
- `dataset`
- `filters`
- `genebank`
- `limit` (integer)
- `order` (string)
- `query`
- `trait`

### `brapi_germplasm_studies` (~182 tokens)

Phenotyping studies (field and glasshouse trials) with trait data for a wheat line, live from the BrAPI nodes
(the INRAE wheat trial network and Whealbi trials in GnpIS (served through FAIDARE) and T3/Wheat, the Triticeae Toolbox (US breeding programmes and regional nurseries)). germplasm: a cultivar or line name (matched exactly; upper/title case are tried too, e.g.
'Soissons' -> 'SOISSONS'). Returns the matching germplasm records (source, holding institute, accession) and each
study's name, trial/network, season, location, number of observations, variables measured and link. Use
brapi_observations for the trait values.

Input parameters:

- `germplasm` (string, required)
- `limit` (integer)

### `brapi_observations` (~230 tokens)

Trait observations for a wheat line from the BrAPI nodes (the INRAE wheat trial network and Whealbi trials in GnpIS (served through FAIDARE) and T3/Wheat, the Triticeae Toolbox (US breeding programmes and regional nurseries)), summarised per study and variable:
n, mean, min, max (numeric values) or the distinct values, season and location. germplasm: line name (as in
brapi_germplasm_studies). trait filters variables by name or ontology ID part (e.g. 'rdt', 'CO_321:1000074',
'yield'; INRAE short names are glossed: rdt = grain yield, prot = grain protein, pmg = thousand-grain weight, ht
\= plant height, rj/rb = yellow/brown rust). study filters by study name part (e.g. 'Le_Moulon', '2015').
Compare values within a study.

Input parameters:

- `germplasm` (string, required)
- `study`
- `trait`

### `wheat_gene_catalogue` (~221 tokens)

Catalogue of Gene Symbols for Wheat (McIntosh et al., GrainGenes; 2024 edition): the official register of
designated wheat genes and alleles. query: a gene or allele symbol ('Lr34', 'Rht-B1b'), locus ('Rht-B1' lists
all its alleles), old synonym ('Rht1', 'Ppd1', 'NAM-B1'), trait words matching a catalogue section ('stem rust',
'height', 'vernalization', 'glutenin') or any text (a cultivar, marker or stock name). chromosome filters by arm
or chromosome ('2D', '7DS'). Each entry gives its section, synonyms, chromosome, the catalogue text (stocks
i:/v:, markers ma:, cloning c:) and, with full, the cited references resolved. Use resolve_gene for the IWGSC
gene ID of cloned genes.

Input parameters:

- `chromosome`
- `full` (boolean)
- `limit` (integer)
- `query` (string, required)

### `kasp_assays` (~358 tokens)

Published KASP assays for wheat genes and SNPs, with primer sequences, alleles, RefSeq v1.0 position and the
allele calls reported for cultivars. Sources: 70 functional assays of Rasheed et al. 2016 (Rht-B1/D1, Ppd-A1/B1/D1,
Vrn-A1/B1/D1, Lr34, Lr21, Lr47, Lr68, Sr2, Yr15, Yr36, Fhb1, Glu-A1/D1, Pina/Pinb, Gpc-B1, TaGW/TGW, PHS, colour
and quality genes; calls in ~300 cultivars) and the CerealsDB KASP core set (960 varietal SNPs with Avalon x
Cadenza map positions; calls in 169 mostly UK varieties). query: gene symbol or synonym (Rht-B1, Rht1, Lr34,
Ppd-D1), v1.1 gene ID, marker/primer name (BS00024039, Vrn1_new) or trait words (vernalization, preharvest
sprouting). region: RefSeq v1.0 interval '4B:30.5-31.5Mb' for assays placed there. line: cultivar name (partial
match) to list its calls (with query: only that gene's assays; alone: all assays scored in that line).
include_calls: lines listed per allele call (0 = counts only when no line is given).

Input parameters:

- `include_calls` (integer)
- `limit` (integer)
- `line`
- `query`
- `region`

### `cereal_trait_genes` (~227 tokens)

Genes cloned in rice for a trait, with their wheat orthologues (A/B/D copies, Ensembl Compara), and rice QTL for
the trait; or, for a wheat or rice gene, what the rice orthologue is known to control. Sources: funRiceGenes (cloned
rice genes with trait keywords, papers and evidence sentences) and Q-TARO (1,051 rice QTL). trait: words such as
'grain size', 'heading date', 'drought', 'blast resistance', 'tiller'. gene: a wheat gene (TraesCS ID or symbol, e.g.
TaGW2) or a rice RAP ID / symbol (Os02g0244100, GW2). Rice gene function in wheat is inferred by orthology: say so,
give the orthology type, and use wheat_mutants / watkins_gene_haplotypes / kasp_assays to test the wheat copies.

Input parameters:

- `gene`
- `limit` (integer)
- `qtl` (boolean)
- `trait`

### `cereal_pangenome_presence` (~216 tokens)

Whether a gene is present and conserved across the barley pangenome (Morex plus 76 cultivars, landraces and wild
barley) and rice assemblies (Nipponbare, 15 Asian rice cultivars, wild Oryza) in Ensembl Plants. gene: a wheat gene
(TraesCS ID or symbol; its barley and rice orthologues from Ensembl Compara are used), a barley Morex ID
(HORVU.MOREX.r3...) or a rice RAP ID. crop: 'barley' or 'rice' (default both). For each reference gene: number of
assemblies with a reciprocal best hit, a one-way hit, or no hit (possible absence or divergence), and the identity
range. assemblies: list up to N assemblies per class. Absence of a hit can be an annotation gap: check the genome
before calling a deletion.

Input parameters:

- `assemblies` (integer)
- `crop`
- `gene` (string, required)

### `editing_candidates` (~246 tokens)

Rank wheat gene-editing targets for a trait: genes cloned in rice for the trait (funRiceGenes, ranked by papers)
whose wheat orthologues have little functional variation in the Watkins landraces and modern lines (so the
variation cannot be selected by crossing), are expressed in the target tissue, and have Kronos EMS truncating
alleles in the A and B copies (testable now as double knockouts). trait: words as for cereal_trait_genes ('grain
size', 'heading date', 'tiller'); tissue: grain, spike, leaf, root or stem (expression filter); candidates: rice
genes examined (most-cited first). Score 0-100 = donor evidence (25) + one-to-one orthology (15) + no common
coding variation in expressed copies (25) + expressed (15) + Kronos truncating lines in A and B (20); each part is
returned. A candidate list to check, not a result.

Input parameters:

- `candidates` (integer)
- `limit` (integer)
- `min_tpm` (number)
- `tissue` (string)
- `trait` (string, required)

### `gwas_associations` (~257 tokens)

Published GWAS marker-trait associations curated by GWAS Atlas (CNCB-NGDC), each with its position, trait
(Plant Trait Ontology), p-value, R2, population and paper (PMID). Data here: wheat (35,430 associations from 96 papers, IWGSC RefSeq v1.0, with the nearest RefSeq v1.1 gene) and rye (3,509 associations from 2 papers, Lo7 = Ensembl secale_cereale). trait: words
('plant height', 'stripe rust', 'grain yield'); region: 'chromosome:start-end' on the stated assembly (e.g.
'6A:8-9Mb'); gene: a gene ID or symbol, giving associations within window_kb of it; species: limit to one species
(Ensembl name); max_p: keep p-values at or below this. Results are sorted by p-value; by_trait and by_paper
summarise all matches.

Input parameters:

- `gene`
- `limit` (integer)
- `max_p`
- `region`
- `species`
- `trait`
- `window_kb` (integer)

### `search_literature` (~246 tokens)

Search publications and preprints in Europe PMC (PubMed, PMC, bioRxiv and others).

query: free text or Europe PMC syntax (e.g. 'grain size QTL 6A', 'TITLE:"pan-genome"').
gene: optional gene symbol or IWGSC ID; its curated synonyms and RefSeq IDs are OR-ed into the query
(e.g. gene='Lr34' also matches Yr18, Sr57, TraesCS7D02G080300).
Words are matched in title and abstract; full_text=True also searches open-access full text.
crop_only requires wheat/Triticum/Aegilops in title or abstract. sort: 'relevance', 'date' (newest first) or 'cited'.
Returns hit count and papers with DOI/PMID, year, journal, open-access flag, citations and a short abstract.

Input parameters:

- `crop_only` (boolean)
- `from_year`
- `full_text` (boolean)
- `gene`
- `limit` (integer)
- `open_access_only` (boolean)
- `query` (string, required)
- `sort` (string)

### `get_paper` (~52 tokens)

Full record for one paper: DOI (10.xxxx/..), PMID (digits) or PMCID (PMC...). Returns the abstract,
open-access status and full-text links.

Input parameters:

- `identifier` (string, required)

### `gene_pathways` (~90 tokens)

Plant Reactome metabolic and signalling pathways of a gene (Gramene), with the top-level category and a link
to the pathway diagram. gene: an Ensembl Plants gene ID or a gene name/symbol. Pathways are projected from rice
by orthology, so say they are inferred. Use pathway_genes for the reverse (all genes of a pathway).

Input parameters:

- `gene` (string, required)

### `pathway_genes` (~139 tokens)

Genes in a Plant Reactome pathway: pathway is a name or words from it ('jasmonic acid biosynthesis',
'gibberellin', 'flavonoid') or a stable ID (R-TAE-1119332). species: a species name as Plant Reactome writes it
('Triticum aestivum', 'Brassica napus', 'Glycine max'); default all species loaded for this site. Matching pathways
are listed with their genes per species. Membership is projected from rice by orthology.

Input parameters:

- `limit` (integer)
- `pathway` (string, required)
- `species`

### `trait_hits` (~165 tokens)

Trait-associated positions across the genome for a chromosome overview: GWAS marker-trait associations and
peaks, named trait genes and mapped trait loci held by this site, with chromosome sizes, each with a genome browser
link where the assembly is loaded. trait: words to match (e.g. 'height', 'oil', 'alkaloid'); chromosome filters
one chromosome. Map marker tracks are left out unless a chromosome is given. Tracks with karyotype false are on
another assembly than chromosome_sizes (pulse: soybean and common bean, Ensembl chromosome numbers; faba bean on
Hedin/2 v2 and projected v1; lentil CDC Redberry v2).

Input parameters:

- `chromosome`
- `limit` (integer)
- `trait`

### `genes_with_annotation` (~235 tokens)

Find genes by function: GO term, InterPro or Pfam domain (accession or words, e.g. 'NB-ARC', 'IPR002182',
'GO:0009414', 'disease resistance', 'MADS-box') or Plant Reactome pathway ('gibberellin biosynthesis',
R-TAE-...; kind='pathway'). kind: GO, InterPro, Pfam or pathway (default: GO + InterPro + Pfam). Optional species
(Ensembl name) and region (chromosome, start, end in bp on the Ensembl/RefSeq v1.0 assembly) filters, e.g. all
NLR genes in a QTL interval. Returns matching genes with location, matched terms and genome browser link, plus
facet counts by species, chromosome and term. Annotations are electronic (InterProScan, GO from InterPro2GO).

Input parameters:

- `chromosome`
- `end`
- `kind`
- `limit` (integer)
- `species`
- `start`
- `term` (string, required)

### `gene_neighbourhood` (~188 tokens)

Gene neighbourhood (microsynteny): the gene and its flanking genes, with their orthologues in related genomes
(wheat homoeologues, Ae. tauschii, durum, barley, Brachypodium, rice) from Ensembl Compara. For each target genome, the collinear block holding most
orthologues of the neighbourhood is reported with how many neighbours are conserved there, so a candidate gene can
be checked for a conserved genomic context. flank: genes each side (default 8, max 25). targets: Ensembl genome
names (default ['aegilops_tauschii', 'triticum_turgidum', 'hordeum_vulgare', 'brachypodium_distachyon', 'oryza_sativa']).

Input parameters:

- `flank` (integer)
- `gene` (string, required)
- `targets`

### `gene_papers` (~96 tokens)

Papers mentioning a gene: Europe PMC search for its gene IDs (all versions, in open-access full text) and its
name/symbols (in title or abstract, with wheat/Triticum/Aegilops). Each paper says whether it matched an ID or a
name. Name matches can be another gene with the same symbol: check the abstract.

Input parameters:

- `gene` (string, required)
- `limit` (integer)

### `annotate_genes` (~146 tokens)

Annotate a list of genes (up to 500 IDs or names) in one call: resolved ID, species, location, name, top
InterPro/Pfam domains, GO terms, Plant Reactome pathways and orthologues in the main related genomes
(['aegilops_tauschii', 'triticum_turgidum', 'hordeum_vulgare', 'brachypodium_distachyon', 'oryza_sativa']). For gene-list follow-up after an experiment (DE genes, GWAS candidates); the web
page genes.html (gene-list tab) runs this and downloads a TSV.

Input parameters:

- `genes` (required)
- `targets`

### `blast_sequence` (~199 tokens)

BLAST a DNA or protein sequence (raw or FASTA; up to 20 kb DNA or 5,000 aa) against this site's databases:
Chinese Spring RefSeq v1.0 genome, RefSeq v1.1 CDS and proteins. The program is chosen from the query and database types (blastn, or blastn-short under 50 nt;
blastx; blastp; tblastn against CDS only). databases: ids from the list this tool returns (default: DNA query ->
reference genome + all CDS sets; protein -> all protein sets). Hits give gene ID (CDS/protein) or genome position
with overlapping genes, identity, alignment length, e-value and a genome browser link. People can use the
SequenceServer page at /blast/ for the same databases.

Input parameters:

- `databases`
- `evalue` (number)
- `max_hits` (integer)
- `sequence` (string, required)

### `wheat_gene_view` (~136 tokens)

Interactive gene view for one bread wheat gene: a relationship map of its A/B/D homoeologues, orthologues
(durum, Ae. tauschii, barley, rice, maize, Arabidopsis and others), EMS knockout lines per copy, Watkins
haplotypes, expression responses and recent papers, with tabs for mutations along the gene and the haplotype
map. Clients that support MCP Apps show it in the chat; the result also lists the nodes and the askWheat gene
page link (page) to share. gene: symbol or any IWGSC ID.

Input parameters:

- `gene` (string, required)

### `genesys_accession` (~233 tokens)

Genebank accessions from Genesys (Crop Trust), live: full passport data (holder, accession number and names,
taxon, biological status, country and site of origin, coordinates, pedigree), whether the accession is available,
in the Multilateral System (SMTA) and safety-duplicated in Svalbard, whether the holder takes requests through
Genesys, a DOI where assigned, and the climate at the collection site (WorldClim: mean temperature, temperature
extremes, precipitation). accession: accession number (e.g. 'PI 591780', 'IG 32990'); holder: FAO WIEWS code
(e.g. USA029, LBN002, AUS165) to make it unique; name: free text (e.g. 'Chinese Spring') when the number is not
known; crop: Genesys crop code (wheat, barley, chickpea, lentil, pea ...) to narrow a name search.

Input parameters:

- `accession`
- `crop`
- `holder`
- `limit` (integer)
- `name`

## Diagnostics

Captured diagnostic sections: TLS, DNSSEC, Authorisation, Transports. The full working is on the page: https://verifymcp.io/servers/org-plantdatabases-askwheat/askwheat#diagnostics

## Score history

- 2026-11-21: 66
- 2026-11-20: 66
- 2026-11-19: 66
- 2026-11-18: 65
- 2026-11-17: 65
- 2026-11-16: 64
- 2026-11-15: 64
- 2026-11-14: 63

## Common questions

### What is the askWheat MCP server?

askWheat is an MCP server listed in the public MCP registry as org.plantdatabases/askwheat. Wheat genes, Watkins pangenome, EMS mutants, expression, genotypes, germplasm and papers (WheatIS). This page covers its hosted endpoint (https://askwheat.plantdatabases.org/mcp).

### Is the askWheat MCP server safe to use?

askWheat scores 66 out of 100 on VerifyMCP. That is a record of what we were able to check automatically, not an endorsement. The category breakdown on this page shows every signal behind the number, including the ones we could not confirm.

### What tools does the askWheat MCP server expose?

askWheat exposes 65 tools: search_data, find_germplasm, data_sources, resolve_gene, gene_function, and 60 more. Their descriptions and schemas cost roughly 11,137 tokens of context every time the server is loaded.

### Does the askWheat MCP server require authentication?

No. We connected to askWheat without credentials and it answered, so anything it exposes is reachable by anyone who knows the address.

### Is the askWheat MCP server still maintained?

askWheat is still listed as active in the MCP registry. We last reached this channel on 21 November 2026. Those dates come from our own scans of the registry and the channel itself, not from anything the publisher announced.

### What did VerifyMCP test on the askWheat MCP server?

VerifyMCP ran 14 checks on askWheat, covering TLS certificate, Authorization, HTTPS, HSTS header and 10 more. Supply Chain Security and Provenance & Transparency do not apply to a hosted endpoint, so they were not tested. The score does not measure everything, and the scoring documentation lists what it leaves out.

## Links

- Remote endpoint: https://askwheat.plantdatabases.org/mcp
- Website: https://askwheat.plantdatabases.org/
- Changelog RSS feed: https://verifymcp.io/servers/org-plantdatabases-askwheat/askwheat.xml
- Changelog JSON feed: https://verifymcp.io/servers/org-plantdatabases-askwheat/askwheat/changelog.json
- Trust report (JSON): https://verifymcp.io/servers/org-plantdatabases-askwheat/askwheat.json
- HTML version of this page: https://verifymcp.io/servers/org-plantdatabases-askwheat/askwheat
