CertScore.ai — Light mode MCP server

Retrieve scan report display content as paginated JSON, without internal diagnostic JSON downloads. The response also offers a single-file full JSON download; private JSON download links expire after five minutes and need no OAuth header; use pagination if your host blocks file downloads. Repeated display records use reportContentRef JSON Pointers. Includes including evidence tables, full-site page and resource inventories, all retained additional-page form fields, form snapshot download references, and retained limitations. Snapshot images are downloaded separately from the returned URLs, with OAuth bearer authentication for workspace scans. Available on OAuth and Light. Start with scanId; follow pagination.nextCursor until complete. Pages share a snapshot; restart if it changes. Each entry has a JSON Pointer path and value; oversized strings use numbered parts. Export completion is not complete observation coverage. Use the concise scan bundle for summaries; use this tool for exhaustive report evidence. No new scan is created.

Returns the completed or completed-limited CertScore evidence bundle for a stable scanId as concise TextContent and matching structuredContent. Available sections include the canonical report overview, bounded projected findings, pre-consent cookie and tracker evidence, coverage limitations, persisted execution provenance, and retrieval URLs. Detail tiers and byte budgets control the bounded response, with explicit returned, total, truncated, and omitted-section metadata. Accept and Reject results distinguish registered decisions from retained after-click facts. Their execution reports succeeded for a completed click and bounded observation, and succeeded_with_confirmation when the consent decision is also verified. Optional afterAction summaries remain useful when registration is unconfirmed; absent or failed capture remains explicitly limited. Consume canonical findings for any scoring effect. Results are automated public-web observations, not legal advice, certification, or a compliance determination.

Returns lifecycle status for a stable CertScore scanId. Active responses include phase, heartbeat, estimated progress, retryAfterSeconds, and sometimes a bounded preliminary preConsentPreview. Terminal responses include completion status, CertScore score and risk metadata when available, coverage, persisted execution region and timestamps, report URL, and a next-action field. Preliminary observations are distinct from completed findings.

Creates a public-website privacy scan or reuses an eligible recent completed scan. Coverage includes pre-consent storage, trackers, consent and CMP signals, privacy-policy disclosures, transport security, and GDPR/ePrivacy or CCPA/CPRA review signals. The response contains a stable scanId, lifecycle status, retry timing, and sometimes a bounded preliminary preConsentPreview; preliminary data contains no final findings or score. Results are automated public-web observations, not legal advice, certification, or a compliance determination. Tool and workflow documentation: https://certscore.ai/developers/mcp.

Call after completed or completed_limited status. summary returns the canonical overview without finding bodies; findings reserves space for compact findings; evidence reserves findings plus bounded evidence digests and references; full adds all available bounded sections. Every response declares detail and byte-budget metadata, omittedSections, retrieval URLs, and nextRecommendedMaxBytes when truncated. Never interpret no-go, not-observed, or limited coverage as proof of compliance.

Poll with only the stable scanId returned by scan_site. Active responses include phase, heartbeat, estimated progress, stalled state, and retry delay. Terminal responses include the canonical score, risk, coverage, timestamps, report URL, and an explicit next action. Stop polling at any terminal status.

First call. Starts or reuses a public-web scan and waits up to 45 seconds by default. If status is queued, running, or finalizing, retain scanId and poll get_scan_status using only that scanId. Stop polling at completed, completed_limited, failed, expired, or rate_limited. For usable completion, call get_scan_bundle. No-go and limited coverage are observations, never proof of compliance.