`dodil k3 vector templates`

Browse the vector-pillar template catalog. Server pins category=embedding — only embedding templates (*_embedding_index pairs) come back.


dodil k3 vector templates [--search TEXT] [--label KEY=VALUE]

The catalog is org-scoped — same templates regardless of which bucket you call from. The -b / --bucket flag is inherited from the parent dodil k3 vector group and accepted but ignored by the server on this subcommand; you can omit it.

Flag	Type	Description
`--search`	string	Free-text match against name + description
`--label`	string list (repeat)	Filter by label `key=value`. Repeat for multiple — all must match (AND).

Examples


# All vector-pillar templates
dodil k3 vector templates -o json | jq '.templates[] | {id, modalities}'
 
# Free-text search
dodil k3 vector templates --search code
 
# Label filter — only image-modality templates
dodil k3 vector templates --label modality=image
 
# Compose — text-modality + ready status
dodil k3 vector templates --label modality=text --label status=ready

--category is not exposed. Vector templates are always category=embedding; nothing to filter on. Use --search and --label to narrow.

The vector catalog at a glance

Each vector template ships as an index + search pair — you create a collection with the _index variant via dodil k3 vector collection add, then the matching _search variant runs query-side automatically.

Template (index)	Modality	Use case
`text_embedding_index`	text, pdf, docx, html, audio, video	Canonical RAG ingest. Chunks + embeds documents.
`code_embedding_index`	code (rust, python, javascript, typescript, go, java, c, cpp, …)	AST-aware chunking (tree-sitter) at function / class boundaries
`visual_embedding_index`	image, video, audio, pdf	Multimodal — frames / spectrograms / page renders
`face_embedding_index`	image	SCRFD detection + face embeddings
`object_embedding_index`	image	Open-vocabulary object detection embeddings (requires `labels` template input — use API)

Full descriptions: Pipelines → Templates → The catalog.

Inspect a template’s contract

To see what inputs a template requires (especially template_inputs for AddVectorPipeline), use the top-level dodil k3 template get:


dodil k3 template get text_embedding_index -o json | jq '{name, description, modalities, contract: .contract.inputs}'

dodil k3 vector templates vs dodil k3 template list vs dodil k3 table templates — same --search / --label filter shape across all three. dodil k3 template list returns all templates across pillars; dodil k3 vector templates filters to category=embedding; dodil k3 table templates filters to warehouse_compatible=true. All three are org-scoped (no -b needed).

Using a template — the canonical flow


# 1. Browse — pick a template
dodil k3 vector templates -o json | jq '.templates[] | {id, modalities, acceptedExtensions}'
 
# 2. Inspect its contract (what inputs it requires)
dodil k3 template get text_embedding_index -o json | jq '.contract.inputs'
 
# 3. Create a collection bound to it
dodil k3 vector collection add docs -b kb-prod \
  --template text_embedding_index \
  --description "RAG corpus"
 
# 4. Upload a document — auto-generated ingest rule fires
dodil k3 object create ./paper.pdf -b kb-prod -k papers/paper.pdf
 
# 5. Search
dodil k3 search "your query" -b kb-prod -c docs

For the full recipe (including the auto-generated pipeline + rule + multi-document ingest), see Recipes → Pipeline Collection.

dodil k3 vector templates

Examples

The vector catalog at a glance

Inspect a template’s contract

Using a template — the canonical flow

See also

`dodil k3 vector templates`