OutSend

Each tool targets a distinct resource or action: creation tools (scrap, enrichment, pipeline, veille) are clearly separated, estimation and retrieval tools (estimate_scrap, get_job, get_job_results, get_events) have unique purposes, and listing tools cover different entity types (jobs, modules, pipelines, veilles). No two tools appear to perform the same function, even the three get_* tools are clearly scoped to status, results, and events.

Tool names follow a consistent verb_noun pattern: create_* for starting actions, get_* for retrieving specific items, list_* for enumerations, and a single estimate_* verb for cost estimation. All names are in lowercase snake_case with no mixing of conventions, making the pattern highly predictable.

With 13 tools, the server is well-scoped for a data extraction and enrichment platform. Each tool addresses a distinct aspect of the lifecycle—creation, estimation, listing, retrieval, monitoring, and schema discovery—without unnecessary redundancy or overwhelming breadth.

The tool surface covers the core workflows: creating jobs (scrap, enrichment, pipeline, veille), estimating costs, retrieving status and results, listing resources, and accessing events. Minor gaps exist, such as no explicit tool for deleting or canceling jobs/pipelines, but these may be outside the intended domain, and the presence of get_events and pipeline_schema fills important monitoring and configuration needs.