kdeps-components-voyageai_rerank
v1.0.1kdeps kdeps-component: voyageai_rerank
set components / shelf workflow
PurposeRerank documents by relevance to a query through the kdeps API
LLM context: ~125 tokens when loaded (prompt or tool definition)
Where it runs
Part of the components set, used in workflow and agent mode. Ready-made workflow steps (sql, browser, email, git, ocr...) a resource calls by name; in agent mode they are also tools.
- a resource names a component (or the model calls it as a tool)
- the component's own resources runcomponents/voyageai_rerank
- its output goes back to the caller
Loaded by: kdeps (agent loop), /settings, kdeps validate, kdeps run, kdeps bundle package, kdeps env, kdeps export k8s, kdeps test
In the code: kdeps (agent loop)yaml.(*Parser).ParseWorkflowyaml.(*Parser).loadAssetComponents
Install
A kdeps asset: installed with the first-run profile that includes its set, or the first time a command needs it, and updated without a new release.
Also installs what it requires, when missing or older: schemas/v1, tools/voyageai_rerank.
Pin a version with kdeps install components/voyageai_rerank@<version>.
Contents
component.yaml
apiVersion: kdeps.io/v1
kind: Component
metadata:
name: voyageai_rerank
description: Rerank documents by relevance to a query through the kdeps API
version: "1.0.1"
interface:
inputs:
- name: documents
type: string
required: true
description: JSON array of document texts to rerank
- name: model
type: string
default: rerank-2
description: VoyageAI rerank model
- name: query
type: string
required: true
description: The search query to rank documents against
- name: top_n
type: number
default: 5
description: Number of top results to return
# body is the ranked JSON array. A missing VOYAGEAI_API_KEY is an error. The HTTP envelope stays inside the call.
output: "{{ output('call').body }}"
resources:
- actionId: call
name: Rerank
httpClient:
method: POST
# In-process host. The HTTP executor calls the handler without dialing.
url: http://kdeps.api/apis/tools.kdeps.io/v1/tools/voyageai_rerank/call
headers:
Content-Type: application/json
failOnError: true
timeout: 60s
data:
documents: "{{ get('voyageai_rerank.documents') }}"
model: "{{ get('voyageai_rerank.model') }}"
query: "{{ get('voyageai_rerank.query') }}"
top_n: "{{ get('voyageai_rerank.top_n') }}"
requires:
schemas/v1: ""
tools/voyageai_rerank: ""
version: 1.0.1
requires:
schemas/v1: ""
tools/voyageai_rerank: ""
files:
component.yaml: |
apiVersion: kdeps.io/v1
kind: Component
metadata:
name: voyageai_rerank
description: Rerank documents by relevance to a query through the kdeps API
version: "1.0.1"
interface:
inputs:
- name: documents
type: string
required: true
description: JSON array of document texts to rerank
- name: model
type: string
default: rerank-2
description: VoyageAI rerank model
- name: query
type: string
required: true
description: The search query to rank documents against
- name: top_n
type: number
default: 5
description: Number of top results to return
# body is the ranked JSON array. A missing VOYAGEAI_API_KEY is an error. The HTTP envelope stays inside the call.
output: "{{ output('call').body }}"
resources:
- actionId: call
name: Rerank
httpClient:
method: POST
# In-process host. The HTTP executor calls the handler without dialing.
url: http://kdeps.api/apis/tools.kdeps.io/v1/tools/voyageai_rerank/call
headers:
Content-Type: application/json
failOnError: true
timeout: 60s
data:
documents: "{{ get('voyageai_rerank.documents') }}"
model: "{{ get('voyageai_rerank.model') }}"
query: "{{ get('voyageai_rerank.query') }}"
top_n: "{{ get('voyageai_rerank.top_n') }}"
Versions
| Version | Published | Status |
|---|---|---|
| 1.0.1 | 10/21/2026 | active |
| 1.0.0 | 10/21/2026 | active |
Details
- Author
- kdeps
- License
- Apache-2.0
- Latest Version
- 1.0.1
- Published
- 10/21/2026