Automatic Video Face Blur MCP: Reversible, Restore Original — NoonAI DIS

Most tools have clear action/resource boundaries: upload vs submit vs poll vs fetch result vs download, and billing tools are separate. However, the high-level encrypt_image/encrypt_video wrappers overlap with the lower-level upload_file + submit_encrypt_job flow, and the job status/result/download trio could be confused without reading descriptions. The detailed descriptions clarify these boundaries, so the set is mostly distinct but not perfectly crisp.

All tool names use snake_case and generally follow a predictable verb_noun pattern: get_job_status, submit_encrypt_job, upload_file, purchase_credits, etc. There are minor semantic variations such as encrypt_image as a high-level wrapper versus submit_encrypt_job, but the naming convention itself is consistent throughout.

The server exposes 15 tools, which is within the typical well-scoped range. The count reflects distinct media workflow steps (upload, submit, poll, fetch, download) plus a separate billing/payment surface, without obvious filler tools.

The core image/video anonymization lifecycle is covered: path and base64 uploads, high- and low-level job submission, status polling, result metadata, and file download. Billing coverage includes status, pricing, constraints, usage, credit purchase, and payment-method registration. Minor gaps exist around canceling/deleting jobs or assets and managing stored payment methods beyond initial registration.