Skip to content
Recued
Menu
← Back to packs

AI Search - Cohere

by recued-core v1 app_pack

API capability pack for bounded Cohere REST API operations against https://api.cohere.com only. Enroll a Cohere API key bearer connection named cohere; no credential is embedded in this manifest. The pack reads API-key status, model metadata, datasets, dataset usage, embed jobs, tokenization results, rerank results, and text embedding responses for semantic search, retrieval, RAG evaluation, and embedding operations. It bundles a rerank review brief, an embedding readiness brief, a scheduled model/dataset operations digest, and approval-gated workflows for creating or canceling one async embed job. Writes are limited to creating one embed job for an existing dataset or canceling one embed job. The pack intentionally excludes streaming-only chat and generate endpoints, deprecated classify/connectors surfaces, OAuth connector authorization, fine-tuning create-update-delete, dataset upload/create/delete, audio transcription uploads, batch create/cancel, model deletion or mutation, arbitrary Cohere API passthrough, and any dynamic outbound callback or tool-execution routing.

Trust & control

What installing this whole pack would let it do. Recued grants these permissions at install — review them there before approving.

Where it runs

Device + server Runs on your device (browser) or your server.

What it's allowed to do

Write Web APIs cohere external change
Read: account.check · No approvalRead: model.search · No approvalRead: model.read · No approvalRead: rerank.create · No approvalRead: embed.create · No approvalRead: tokenize.create · No approvalRead: detokenize.create · No approvalRead: dataset.search · No approvalRead: dataset.usage.read · No approvalRead: dataset.read · No approvalRead: embed_job.search · No approvalRead: embed_job.read · No approvalWrite: embed_job.create · Asks approvalWrite: embed_job.cancel · Asks approval

Data it touches

SurfacesWeb APIs

About

Tags

pack:coherecohereaisemantic-searchrerankembeddingsmodelsdatasetsragdeveloper-toolsworkflowrecipeapiv3composition