Skip to main content

Documentation Index

Fetch the complete documentation index at: https://docs.getclaro.ai/llms.txt

Use this file to discover all available pages before exploring further.

The product is organized into three groups in the sidebar: 
CORE          — the persistent catalog and its global views
MODULES       — the operational stages: Onboard, Analyse, Monitor, Distribute
RESEARCH AGENTS — one-off list-building and extraction tasks

​
Core

​
Dashboard

Entry point for any new workflow. The Dashboard splits the platform’s two modes:
  • Catalog Operations — for working against your persistent catalog (onboard SKUs, improve taxonomies, monitor competition, analyze inventory).
  • Research Agents — for one-off lists and extractions (find prospects, parse documents, enrich spreadsheets, scrape URLs).
Recent activity, active operations, and pending review items also surface here.

​
Catalogue

The Catalogue page is where you spend most of your time. Each catalog object opens with five tabs:
TabPurpose
AttributesDefine and edit the schema. Add fields, change types, set validation rules, mark fields as required, set enum values, configure reference relationships.
RecordsThe data itself. Sortable, filterable, editable in a spreadsheet-like grid. Inline edits create change events with provenance.
Data SourceConnect upstream — file uploads, scheduled scrapes, supplier portals, APIs, database connections. Each source has its own mapping to attributes.
OperationRun and schedule operations against the records. See Operations.
ConfigAccess control, retention, write-back targets, default reviewers, auto-apply thresholds.
See the dedicated Catalogue guide for details.

​
Taxonomy

Manage hierarchical categories used across your catalogue. View the tree, edit nodes, version the taxonomy, and track which records map to which leaf. The Generate Taxonomy operation (under Modules → Analyse) can propose a taxonomy from your data; the Taxonomy page is where you curate and publish the result.

​
Similarity & Duplicate

Inspect the similarity graph. Browse clusters, review proposed merges, set match thresholds, and configure which fields drive similarity. This surface is shared by the Find Duplicates and Find Similarities operations. See Similarity & Duplicate.

​
Notifications

Every operation that needs review lands here. Review items are grouped by operation and can be approved, rejected, or edited inline. Bulk actions are available per cluster. See Notifications.

​
Modules

Modules group operations by the stage of work they support. Each module exposes its own surfaces in the sidebar but ultimately reads from and writes to the Catalogue.

​
Onboard

Bringing new data — suppliers, SKUs, files — into the catalogue.
  • Status — pipeline view of in-flight onboarding jobs (file received → mapped → resolved → enriched → ready). Errors and stalls surface here.
  • Supplier Portal — a hosted intake surface you can share with suppliers. Suppliers upload files or fill structured forms; submissions land as Data Sources on the relevant catalogue, pre-mapped where possible.
See Onboard.

​
Analyse

Operations that read the catalogue and produce assessments, suggestions, or transformations.
  • Analyse — a data-quality assessment with per-field scores, distribution stats, critical issues, and an overall Data Quality Score.
  • Generative Engine — model-driven generation: descriptions, marketing copy, attribute extraction, alt-text, translations.
See Analyse.

​
Monitor

Continuous observation operations.
  • Price — track prices across a configurable cadence; flag drifts, deltas, anomalies.
  • Competitor — pull competitor listings for matched SKUs and surface side-by-side comparisons.
Both produce recurring runs and feed alerts into Notifications. See Monitor.

​
Distribute

Pushing clean data downstream.
  • Unified Catalog — a denormalized, query-ready view of the catalogue, available as a read API and as exports.
  • Sync & Export — connectors to Shopify, Amazon, BigQuery, Google Sheets, Webhooks, and S3. Each connector has its own mapping, schedule, and conflict policy.
See Distribute.

​
Research Agents

Task-specific tools for ad-hoc work that doesn’t (yet) belong in your persistent catalogue.
AgentUse case
Find your perfect listGenerate a researched dataset of prospects, suppliers, partners, or any entity class.
Turn documents into structured dataExtract structured records from PDFs, contracts, datasheets, or reports.
Analyze & enrich spreadsheetsUpload a spreadsheet; validate, standardize, and complete missing data.
Scrape data from URLsCollect structured records from web pages.
Outputs land in Generated Datasets and Uploaded Files. Any generated dataset can be promoted into a Catalogue, at which point it becomes a persistent object. A growing library of additional agents is available in Agent Library. See Research Agents.