The Complete export collects the selected stored website-scan evidence tables, including full Lighthouse accessibility audits, media, links, page summaries, sitemap/robots observations, scan diagnostics, and Siteimp's generated read-model analysis.

Format: siteimp-complete-evidence-v1
Primary roots: tables, generated_analysis
Included tables: 20 currently defined website-evidence tables. Raw artifact files are not embedded.

Common export metadata

All five report formats include the same core envelope. Individual files are generated independently from the scan database; they are not slices of another downloaded JSON file.

PropertyMeaning
export_formatExact format identifier; use it when selecting a parser.
export_typeHuman-readable machine-friendly profile name.
purposeShort description of the intended use.
snapshot_idUUID for the scan represented by this report. Match it when combining scan-specific exports.
websiteWebsite identity and starting URL.
exported_at_utcTime the JSON file was generated, not the time the scan ran.
schema_fingerprintNoncryptographic fnv1a64-... identifier for the database table definitions. It is not a hash of scan results.
related_exportsNames and descriptions of the other available profiles, plus instructions for downloading them. It does not embed the other reports.
export_policyScope, retained evidence, and known omissions for the selected profile.

website contains website_key, website_slug, name, start_url, and registrable_domain. related_exports.available maps each profile name to export_format, purpose, and selected; selected identifies the current report, not a report that has already been downloaded.

JSON text stored in database columns ending in _json is decoded into JSON arrays or objects in table-based exports. Do not parse those fields as JSON strings again. SQL NULL becomes JSON null. Integer columns can represent boolean-style flags (0 and 1); consult the Schema export's create_sql for their constraints. Neither object-property ordering nor numeric row IDs should be used as a durable cross-scan identity.

Top-level contract

Complete contains the common header plus:

PropertyShapeMeaning
manifestobjectExact table inclusion list, excluded execution tables, and JSON decoding/artifact policy.
tablesobjectPer-table arrays containing full exported rows.
row_countsobjectNumber of rows written for each included table.
generated_analysisobjectExisting Siteimp Signals and Issues read-model output, not an AI-generated interpretation.

The manifest.included_tables array is the most reliable way to enumerate which tables this version of the profile exports. The current list is documented below. A table with zero records remains present with an empty array.

Complete evidence table inventory

TableEvidence provided
snapshotsStatus, start/completion times, mode, pipeline version, stop reason, scan summary, and failure/timing totals.
artifactsArtifact metadata: ID, kind, page or media references, relative path, checksum, file size, storage and missing flags.
robots_txt_recordsRobots fetch, parse information, body, and checksums.
scan_sitemapsSitemap fetch and URL-set comparison summary.
sitemap_entriesIndividual stored sitemap entry URLs, when present.
sitemap_mismatchesSitemap-only and crawl-only URL discrepancy records.
pagesPage identity, fetch result, HTTP status, redirects, blocking and artifact references.
page_discoveryDiscovered URL, method, source page and discovery time.
linksInternal-link edges and attributes.
external_targetsNormalized external target URLs and hosts.
external_linksExternal-link usage, source page, anchor text, and target references.
external_link_checksFetch/check outcomes for external target IDs.
lighthouse_reportsPage scores, individual accessibility audits, metrics and audit summaries.
scan_accessibility_summarySitewide accessibility coverage, scores, audit totals and top issues.
scan_accessibility_issuesAggregated individual audit findings and per-page supporting evidence.
media_assetsMedia URLs, fetch responses, content types, dimensions, hashes and metadata.
page_mediaMedia use on pages, usage kinds, selectors, URL references, sizes and attributes.
page_html_summaryTitles, descriptions, headings, language, image counts, word counts and HTML-derived metadata.
media_exif_summarySummary of captured EXIF information and identified metadata risks.
scan_diagnosticsScan-wide diagnostics, policies, delays, durations and related observability.

Four scanner execution-state tables are excluded: page_jobs, external_link_jobs, media_jobs, and external_host_state. Their definitions do appear in Schema, but they are not part of this shareable evidence document. Neither raw HTML files nor raw Lighthouse report files nor downloaded media bytes are embedded.

generated_analysis contract

The Complete report also includes Siteimp's existing generated analysis read models, serialized under generated_analysis. This is not a prompt response from an LLM.

FieldMeaning
formatRead-model identifier, currently siteimp-generated-analysis-v1.
snapshot_idScan identity repeated inside this generated analysis object.
signalsSitewide counts, broken link/image signals, external-link concentration, sitemap mismatch.
page_issue_summariesStored read-model summary of page-level issues.
page_issuesArray of per-page issue objects.

A page_issues[] object includes page_key, fetch, broken_internal_links, broken_external_links, broken_images, metadata, and sitemap. These structures reflect the read-model serializers; individual nested fields can evolve independently of the SQLite columns. For a first-pass explanation, the Introductory report is intentionally much smaller.

Relationships and joining rows

  • snapshot_id identifies the scan represented throughout the export.
  • page_key connects pages to Lighthouse reports, HTML summaries, page discovery, media usage, and link source records within that scan.
  • external_target_id relates external_links and external_link_checks to external_targets.id.
  • media_asset_id relates page_media and media_exif_summary to media_assets.id.
  • source_artifact_id or raw_html_artifact_id can refer to artifacts.id; a reference to an artifact is not the artifact file content.
  • For exact foreign keys, constraints and declared column types, consult Schema.

Prefer composite scan-aware keys when combining multiple snapshots. Database-generated integers can be reused or change when data is recreated and should not become cross-scan public IDs.

Local path handling and portability

Complete can contain references to locally stored artifacts. Its export policy converts recognized scan artifact-cache paths to paths relative to the scan's artifact directory. Other absolute local filesystem paths are redacted, including those found inside decoded media metadata and EXIF structures. The original SQLite values are left unchanged.

A relative artifacts[].path is a descriptive reference, not a promise that the referenced file is available to a recipient. Check storage_tier, missing, size_bytes, and sha256 to interpret the stored artifact metadata. No file contents are embedded.

This path transformation is intentional: Complete's tables data may differ from the raw SQLite source for local path fields. For other table values, the current export retains full row fields and decoded JSON text, including N/A audits and bookkeeping timestamps.

Before sharing, consider URLs, extracted page metadata, diagnostics, robots body text, and EXIF observations. Path redaction is not a blanket guarantee that the entire report contains no confidential information.

Example: enumerate evidence and row counts

import json

with open("complete.json", encoding="utf-8") as file:
    report = json.load(file)

assert report["export_format"] == "siteimp-complete-evidence-v1"
for table_name in report["manifest"]["included_tables"]:
    expected = report["row_counts"][table_name]
    actual = len(report["tables"][table_name])
    assert expected == actual, table_name
    print(table_name, actual)

This checks internal counts for this downloaded file. It does not compare with SQLite or prove that the scanner collected every possible kind of evidence.

Why use specialized exports instead?

Complete is deliberately comprehensive, not optimized for a first AI conversation or a narrow task. Accessibility preserves the full relevant audits in a smaller file, Page Discovery isolates crawl provenance, and Introductory presents prioritized findings. Schema describes table definitions separately and may be reusable across many scans with the same fingerprint.

For long-running analysis or dashboards, choose the smallest profile that actually contains the data your code needs. Match snapshot_id across files from one scan and check schema_fingerprint and export_format as separate compatibility indicators.