# Verify ## List Verify Jobs **get** `/api/alpha/verify` List Verify jobs with optional filtering and pagination. Filter by `status`, specific `job_ids`, or creation date range. ### Query Parameters - `created_at_on_or_after: optional string` Include items created at or after this timestamp (inclusive) - `created_at_on_or_before: optional string` Include items created at or before this timestamp (inclusive) - `expand: optional array of string` Optional fields to include (e.g. `result`). - `job_ids: optional array of string` Filter by specific job IDs - `organization_id: optional string` - `page_size: optional number` Number of items per page - `page_token: optional string` Token for pagination - `project_id: optional string` - `status: optional "CANCELLED" or "COMPLETED" or "FAILED" or 2 more` Filter by job status - `"CANCELLED"` - `"COMPLETED"` - `"FAILED"` - `"PENDING"` - `"RUNNING"` ### Cookie Parameters - `session: optional string` ### Returns - `items: array of object { id, configuration, document_input_type, 9 more }` The list of items. - `id: string` Unique identifier - `configuration: object { target_pages, tier }` Verify configuration used for this job - `target_pages: optional string` Comma-separated page numbers or ranges to analyze (1-based). Omit to analyze all pages. Ignored for non-PDF inputs. - `tier: optional "agentic" or "fast"` Verify tier: 'fast' runs only the quick deterministic forensic checks (metadata, content integrity, container structure, pixel statistics); 'agentic' (default) runs the full pipeline including the learned detectors and the semantic review pass. - `"agentic"` - `"fast"` - `document_input_type: "file_id" or "parse_job_id" or "url"` Type of the document input (FILE) - `"file_id"` - `"parse_job_id"` - `"url"` - `file_input: string` ID of the input file - `project_id: string` Project this job belongs to - `status: "CANCELLED" or "COMPLETED" or "FAILED" or 2 more` Current job status: PENDING, RUNNING, COMPLETED, FAILED, or CANCELLED - `"CANCELLED"` - `"COMPLETED"` - `"FAILED"` - `"PENDING"` - `"RUNNING"` - `user_id: string` User who created this job - `created_at: optional string` Creation datetime - `error_message: optional string` Error message if job failed - `result: optional object { detector_version, overall_score, verdict, 9 more }` Result of a Verify (doctored-document) analysis. Raw per-signal detail (evidence list, per-family sub-scores, raw regions, forensic heatmaps) is available separately via the job's details endpoint. - `detector_version: string` Version of the detector that produced the result - `overall_score: number` Overall doctoring likelihood (0 to 1) - `verdict: "AUTHENTIC" or "DOCTORED" or "LIKELY_DOCTORED" or 2 more` Overall verdict for the document - `"AUTHENTIC"` - `"DOCTORED"` - `"LIKELY_DOCTORED"` - `"NO_STRONG_SIGNAL"` - `"SUSPICIOUS"` - `composite_scores: optional object { ai_generated, document_coherence, document_metadata, 3 more }` Composite scores, each answering one question about the document - `ai_generated: optional object { applicable, score }` Was this content synthesized by a generative model? - `applicable: optional boolean` Whether the checks feeding this composite ran on this document. When false the document was not checked for this — not cleared of it - `score: optional number` Score (0 to 1); null when the composite was not applicable - `document_coherence: optional object { applicable, score }` Does the document's content agree with itself (checksums, arithmetic, machine-readable zones)? - `applicable: optional boolean` Whether the checks feeding this composite ran on this document. When false the document was not checked for this — not cleared of it - `score: optional number` Score (0 to 1); null when the composite was not applicable - `document_metadata: optional object { applicable, score }` Does the file's provenance / toolchain history look suspicious? Advisory: individually weak workflow-hygiene signals - `applicable: optional boolean` Whether the checks feeding this composite ran on this document. When false the document was not checked for this — not cleared of it - `score: optional number` Score (0 to 1); null when the composite was not applicable - `known_fraud: optional object { applicable, score }` Has this asset (or its template) been seen in fraud before? - `applicable: optional boolean` Whether the checks feeding this composite ran on this document. When false the document was not checked for this — not cleared of it - `score: optional number` Score (0 to 1); null when the composite was not applicable - `manually_edited: optional object { applicable, score }` Was this document altered after creation (splice, retype, redact, inpaint)? - `applicable: optional boolean` Whether the checks feeding this composite ran on this document. When false the document was not checked for this — not cleared of it - `score: optional number` Score (0 to 1); null when the composite was not applicable - `recapture: optional object { applicable, score }` Was the document captured through a channel that destroys forensic evidence (photo of a screen, print-then-rescan)? - `applicable: optional boolean` Whether the checks feeding this composite ran on this document. When false the document was not checked for this — not cleared of it - `score: optional number` Score (0 to 1); null when the composite was not applicable - `confidence: optional number` Confidence in the verdict (0 to 1): how firmly the detected signals support the verdict bucket, independent of the doctoring likelihood itself - `error: optional string` Error detail when the analysis could not complete - `page_count: optional number` Number of analysed pages (1 for images/docx) - `page_dimensions: optional array of object { height, page, width }` Rendered pixel size per page, so region bboxes can be scaled onto the page - `height: number` Rendered page height in pixels - `page: number` 0-based page index (0 for standalone images) - `width: number` Rendered page width in pixels - `reasoning: optional string` Explanation of the verdict - `suspect_regions: optional array of object { bbox, explanation, kind, 5 more }` Regions that led to the suspected fraud, ranked most-suspect first, each with an explanation of what makes it suspect - `bbox: array of number` Region bounding box as [x, y, w, h] in page-render pixels - `explanation: string` Human-readable explanation of what makes this region suspect - `kind: string` Kind of anomaly detected in this region - `page: number` 0-based page index (0 for standalone images) - `score: number` Suspicion score for this region (0 to 1) - `source: string` Detector that flagged this region - `primary: optional boolean` Whether this region is part of the small set of decisive evidence behind the verdict — the boxes a reviewer should look at first - `review: optional string` Automated reviewer verdict for this region (confirmed, dismissed, unsure, or empty). A dismissed region can still be surfaced when it is the only place to look; this label says how to read it - `synthetic_score: optional number` Likelihood (0 to 1) that the document is wholly generated or fabricated rather than a capture of a real document. Null for jobs completed before this score was introduced - `tampering_score: optional number` Likelihood (0 to 1) that a real captured document was locally edited — a genuine capture with regions altered after the fact. Null for jobs completed before this score was introduced - `transaction_id: optional string` Idempotency key - `updated_at: optional string` Update datetime - `next_page_token: optional string` A token, which can be sent as page_token to retrieve the next page. If this field is omitted, there are no subsequent pages. - `total_size: optional number` The total number of items available. This is only populated when specifically requested. The value may be an estimate and can be used for display purposes only. ### Example ```http curl https://api.cloud.llamaindex.ai/api/alpha/verify \ -H "Authorization: Bearer $LLAMA_CLOUD_API_KEY" ``` #### Response ```json { "items": [ { "id": "id", "configuration": { "target_pages": "1,3,5-7", "tier": "agentic" }, "document_input_type": "file_id", "file_input": "file_input", "project_id": "project_id", "status": "CANCELLED", "user_id": "user_id", "created_at": "2019-12-27T18:11:19.117Z", "error_message": "error_message", "result": { "detector_version": "detector_version", "overall_score": 0, "verdict": "AUTHENTIC", "composite_scores": { "ai_generated": { "applicable": true, "score": 0 }, "document_coherence": { "applicable": true, "score": 0 }, "document_metadata": { "applicable": true, "score": 0 }, "known_fraud": { "applicable": true, "score": 0 }, "manually_edited": { "applicable": true, "score": 0 }, "recapture": { "applicable": true, "score": 0 } }, "confidence": 0, "error": "error", "page_count": 1, "page_dimensions": [ { "height": 0, "page": 0, "width": 0 } ], "reasoning": "reasoning", "suspect_regions": [ { "bbox": [ 0, 0, 0, 0 ], "explanation": "explanation", "kind": "kind", "page": 0, "score": 0, "source": "source", "primary": true, "review": "review" } ], "synthetic_score": 0, "tampering_score": 0 }, "transaction_id": "transaction_id", "updated_at": "2019-12-27T18:11:19.117Z" } ], "next_page_token": "next_page_token", "total_size": 0 } ``` ## Create Verify Job **post** `/api/alpha/verify` Create a Verify job. Analyzes a document for signs of doctoring (splicing, copy-move, AI generation, metadata tampering, ...). Set `file_input` to a file ID (`dfl-...`). Optionally provide a `configuration` object to control the semantic agent. The job runs asynchronously. Poll `GET /verify/{job_id}` with `expand=result` to check status and retrieve results. ### Query Parameters - `organization_id: optional string` - `project_id: optional string` ### Cookie Parameters - `session: optional string` ### Body Parameters - `configuration: optional object { target_pages, tier }` Configuration for a Verify job. - `target_pages: optional string` Comma-separated page numbers or ranges to analyze (1-based). Omit to analyze all pages. Ignored for non-PDF inputs. - `tier: optional "agentic" or "fast"` Verify tier: 'fast' runs only the quick deterministic forensic checks (metadata, content integrity, container structure, pixel statistics); 'agentic' (default) runs the full pipeline including the learned detectors and the semantic review pass. - `"agentic"` - `"fast"` - `file_id: optional string` Deprecated: use file_input instead - `file_input: optional string` File ID of the document to analyze - `transaction_id: optional string` Idempotency key scoped to the project. Reusing a key returns the original job; the new request body is ignored. - `webhook_configuration_ids: optional array of string` IDs of saved webhook configurations to notify for this job. - `webhook_configurations: optional array of object { webhook_events, webhook_headers, webhook_output_format, 2 more }` Outbound webhook endpoints to notify on job status changes - `webhook_events: optional array of "batch.cancelled" or "batch.error" or "batch.pending" or 35 more` Events to subscribe to (e.g. 'parse.success', 'extract.error'). If null, all events are delivered. - `"batch.cancelled"` - `"batch.error"` - `"batch.pending"` - `"batch.running"` - `"batch.success"` - `"classify.cancelled"` - `"classify.error"` - `"classify.partial_success"` - `"classify.pending"` - `"classify.running"` - `"classify.success"` - `"extract.cancelled"` - `"extract.error"` - `"extract.partial_success"` - `"extract.pending"` - `"extract.success"` - `"parse.cancelled"` - `"parse.error"` - `"parse.partial_success"` - `"parse.pending"` - `"parse.running"` - `"parse.success"` - `"sheets.cancelled"` - `"sheets.error"` - `"sheets.partial_success"` - `"sheets.pending"` - `"sheets.success"` - `"split.cancelled"` - `"split.error"` - `"split.pending"` - `"split.processing"` - `"split.success"` - `"unmapped_event"` - `"verify.cancelled"` - `"verify.error"` - `"verify.pending"` - `"verify.running"` - `"verify.success"` - `webhook_headers: optional map[string]` Custom HTTP headers sent with each webhook request (e.g. auth tokens) - `webhook_output_format: optional string` Response format sent to the webhook: 'string' (default) or 'json' - `webhook_signing_secret: optional string` Shared signing secret used to sign webhook deliveries. When set, each request includes an HMAC-SHA256 signature of the request body in the 'LC-Signature' header (value 'sha256='). Recompute the HMAC over the raw request body with this secret to verify the delivery is authentic. - `webhook_url: optional string` URL to receive webhook POST notifications ### Returns - `id: string` Unique identifier - `configuration: object { target_pages, tier }` Verify configuration used for this job - `target_pages: optional string` Comma-separated page numbers or ranges to analyze (1-based). Omit to analyze all pages. Ignored for non-PDF inputs. - `tier: optional "agentic" or "fast"` Verify tier: 'fast' runs only the quick deterministic forensic checks (metadata, content integrity, container structure, pixel statistics); 'agentic' (default) runs the full pipeline including the learned detectors and the semantic review pass. - `"agentic"` - `"fast"` - `document_input_type: "file_id" or "parse_job_id" or "url"` Type of the document input (FILE) - `"file_id"` - `"parse_job_id"` - `"url"` - `file_input: string` ID of the input file - `project_id: string` Project this job belongs to - `status: "CANCELLED" or "COMPLETED" or "FAILED" or 2 more` Current job status: PENDING, RUNNING, COMPLETED, FAILED, or CANCELLED - `"CANCELLED"` - `"COMPLETED"` - `"FAILED"` - `"PENDING"` - `"RUNNING"` - `user_id: string` User who created this job - `created_at: optional string` Creation datetime - `error_message: optional string` Error message if job failed - `result: optional object { detector_version, overall_score, verdict, 9 more }` Result of a Verify (doctored-document) analysis. Raw per-signal detail (evidence list, per-family sub-scores, raw regions, forensic heatmaps) is available separately via the job's details endpoint. - `detector_version: string` Version of the detector that produced the result - `overall_score: number` Overall doctoring likelihood (0 to 1) - `verdict: "AUTHENTIC" or "DOCTORED" or "LIKELY_DOCTORED" or 2 more` Overall verdict for the document - `"AUTHENTIC"` - `"DOCTORED"` - `"LIKELY_DOCTORED"` - `"NO_STRONG_SIGNAL"` - `"SUSPICIOUS"` - `composite_scores: optional object { ai_generated, document_coherence, document_metadata, 3 more }` Composite scores, each answering one question about the document - `ai_generated: optional object { applicable, score }` Was this content synthesized by a generative model? - `applicable: optional boolean` Whether the checks feeding this composite ran on this document. When false the document was not checked for this — not cleared of it - `score: optional number` Score (0 to 1); null when the composite was not applicable - `document_coherence: optional object { applicable, score }` Does the document's content agree with itself (checksums, arithmetic, machine-readable zones)? - `applicable: optional boolean` Whether the checks feeding this composite ran on this document. When false the document was not checked for this — not cleared of it - `score: optional number` Score (0 to 1); null when the composite was not applicable - `document_metadata: optional object { applicable, score }` Does the file's provenance / toolchain history look suspicious? Advisory: individually weak workflow-hygiene signals - `applicable: optional boolean` Whether the checks feeding this composite ran on this document. When false the document was not checked for this — not cleared of it - `score: optional number` Score (0 to 1); null when the composite was not applicable - `known_fraud: optional object { applicable, score }` Has this asset (or its template) been seen in fraud before? - `applicable: optional boolean` Whether the checks feeding this composite ran on this document. When false the document was not checked for this — not cleared of it - `score: optional number` Score (0 to 1); null when the composite was not applicable - `manually_edited: optional object { applicable, score }` Was this document altered after creation (splice, retype, redact, inpaint)? - `applicable: optional boolean` Whether the checks feeding this composite ran on this document. When false the document was not checked for this — not cleared of it - `score: optional number` Score (0 to 1); null when the composite was not applicable - `recapture: optional object { applicable, score }` Was the document captured through a channel that destroys forensic evidence (photo of a screen, print-then-rescan)? - `applicable: optional boolean` Whether the checks feeding this composite ran on this document. When false the document was not checked for this — not cleared of it - `score: optional number` Score (0 to 1); null when the composite was not applicable - `confidence: optional number` Confidence in the verdict (0 to 1): how firmly the detected signals support the verdict bucket, independent of the doctoring likelihood itself - `error: optional string` Error detail when the analysis could not complete - `page_count: optional number` Number of analysed pages (1 for images/docx) - `page_dimensions: optional array of object { height, page, width }` Rendered pixel size per page, so region bboxes can be scaled onto the page - `height: number` Rendered page height in pixels - `page: number` 0-based page index (0 for standalone images) - `width: number` Rendered page width in pixels - `reasoning: optional string` Explanation of the verdict - `suspect_regions: optional array of object { bbox, explanation, kind, 5 more }` Regions that led to the suspected fraud, ranked most-suspect first, each with an explanation of what makes it suspect - `bbox: array of number` Region bounding box as [x, y, w, h] in page-render pixels - `explanation: string` Human-readable explanation of what makes this region suspect - `kind: string` Kind of anomaly detected in this region - `page: number` 0-based page index (0 for standalone images) - `score: number` Suspicion score for this region (0 to 1) - `source: string` Detector that flagged this region - `primary: optional boolean` Whether this region is part of the small set of decisive evidence behind the verdict — the boxes a reviewer should look at first - `review: optional string` Automated reviewer verdict for this region (confirmed, dismissed, unsure, or empty). A dismissed region can still be surfaced when it is the only place to look; this label says how to read it - `synthetic_score: optional number` Likelihood (0 to 1) that the document is wholly generated or fabricated rather than a capture of a real document. Null for jobs completed before this score was introduced - `tampering_score: optional number` Likelihood (0 to 1) that a real captured document was locally edited — a genuine capture with regions altered after the fact. Null for jobs completed before this score was introduced - `transaction_id: optional string` Idempotency key - `updated_at: optional string` Update datetime ### Example ```http curl https://api.cloud.llamaindex.ai/api/alpha/verify \ -H 'Content-Type: application/json' \ -H "Authorization: Bearer $LLAMA_CLOUD_API_KEY" \ -d '{}' ``` #### Response ```json { "id": "id", "configuration": { "target_pages": "1,3,5-7", "tier": "agentic" }, "document_input_type": "file_id", "file_input": "file_input", "project_id": "project_id", "status": "CANCELLED", "user_id": "user_id", "created_at": "2019-12-27T18:11:19.117Z", "error_message": "error_message", "result": { "detector_version": "detector_version", "overall_score": 0, "verdict": "AUTHENTIC", "composite_scores": { "ai_generated": { "applicable": true, "score": 0 }, "document_coherence": { "applicable": true, "score": 0 }, "document_metadata": { "applicable": true, "score": 0 }, "known_fraud": { "applicable": true, "score": 0 }, "manually_edited": { "applicable": true, "score": 0 }, "recapture": { "applicable": true, "score": 0 } }, "confidence": 0, "error": "error", "page_count": 1, "page_dimensions": [ { "height": 0, "page": 0, "width": 0 } ], "reasoning": "reasoning", "suspect_regions": [ { "bbox": [ 0, 0, 0, 0 ], "explanation": "explanation", "kind": "kind", "page": 0, "score": 0, "source": "source", "primary": true, "review": "review" } ], "synthetic_score": 0, "tampering_score": 0 }, "transaction_id": "transaction_id", "updated_at": "2019-12-27T18:11:19.117Z" } ``` ## Get Verify Job **get** `/api/alpha/verify/{job_id}` Get a Verify job by ID. Returns the job status and configuration. Pass `expand=result` to include the Verify result (overall score, verdict, confidence, composite scores, and suspect regions) when the job is complete. Raw per-signal detail is available via `GET /verify/{job_id}/details`. ### Path Parameters - `job_id: string` ### Query Parameters - `expand: optional array of string` Optional fields to include (e.g. `result`). - `organization_id: optional string` - `project_id: optional string` ### Cookie Parameters - `session: optional string` ### Returns - `id: string` Unique identifier - `configuration: object { target_pages, tier }` Verify configuration used for this job - `target_pages: optional string` Comma-separated page numbers or ranges to analyze (1-based). Omit to analyze all pages. Ignored for non-PDF inputs. - `tier: optional "agentic" or "fast"` Verify tier: 'fast' runs only the quick deterministic forensic checks (metadata, content integrity, container structure, pixel statistics); 'agentic' (default) runs the full pipeline including the learned detectors and the semantic review pass. - `"agentic"` - `"fast"` - `document_input_type: "file_id" or "parse_job_id" or "url"` Type of the document input (FILE) - `"file_id"` - `"parse_job_id"` - `"url"` - `file_input: string` ID of the input file - `project_id: string` Project this job belongs to - `status: "CANCELLED" or "COMPLETED" or "FAILED" or 2 more` Current job status: PENDING, RUNNING, COMPLETED, FAILED, or CANCELLED - `"CANCELLED"` - `"COMPLETED"` - `"FAILED"` - `"PENDING"` - `"RUNNING"` - `user_id: string` User who created this job - `created_at: optional string` Creation datetime - `error_message: optional string` Error message if job failed - `result: optional object { detector_version, overall_score, verdict, 9 more }` Result of a Verify (doctored-document) analysis. Raw per-signal detail (evidence list, per-family sub-scores, raw regions, forensic heatmaps) is available separately via the job's details endpoint. - `detector_version: string` Version of the detector that produced the result - `overall_score: number` Overall doctoring likelihood (0 to 1) - `verdict: "AUTHENTIC" or "DOCTORED" or "LIKELY_DOCTORED" or 2 more` Overall verdict for the document - `"AUTHENTIC"` - `"DOCTORED"` - `"LIKELY_DOCTORED"` - `"NO_STRONG_SIGNAL"` - `"SUSPICIOUS"` - `composite_scores: optional object { ai_generated, document_coherence, document_metadata, 3 more }` Composite scores, each answering one question about the document - `ai_generated: optional object { applicable, score }` Was this content synthesized by a generative model? - `applicable: optional boolean` Whether the checks feeding this composite ran on this document. When false the document was not checked for this — not cleared of it - `score: optional number` Score (0 to 1); null when the composite was not applicable - `document_coherence: optional object { applicable, score }` Does the document's content agree with itself (checksums, arithmetic, machine-readable zones)? - `applicable: optional boolean` Whether the checks feeding this composite ran on this document. When false the document was not checked for this — not cleared of it - `score: optional number` Score (0 to 1); null when the composite was not applicable - `document_metadata: optional object { applicable, score }` Does the file's provenance / toolchain history look suspicious? Advisory: individually weak workflow-hygiene signals - `applicable: optional boolean` Whether the checks feeding this composite ran on this document. When false the document was not checked for this — not cleared of it - `score: optional number` Score (0 to 1); null when the composite was not applicable - `known_fraud: optional object { applicable, score }` Has this asset (or its template) been seen in fraud before? - `applicable: optional boolean` Whether the checks feeding this composite ran on this document. When false the document was not checked for this — not cleared of it - `score: optional number` Score (0 to 1); null when the composite was not applicable - `manually_edited: optional object { applicable, score }` Was this document altered after creation (splice, retype, redact, inpaint)? - `applicable: optional boolean` Whether the checks feeding this composite ran on this document. When false the document was not checked for this — not cleared of it - `score: optional number` Score (0 to 1); null when the composite was not applicable - `recapture: optional object { applicable, score }` Was the document captured through a channel that destroys forensic evidence (photo of a screen, print-then-rescan)? - `applicable: optional boolean` Whether the checks feeding this composite ran on this document. When false the document was not checked for this — not cleared of it - `score: optional number` Score (0 to 1); null when the composite was not applicable - `confidence: optional number` Confidence in the verdict (0 to 1): how firmly the detected signals support the verdict bucket, independent of the doctoring likelihood itself - `error: optional string` Error detail when the analysis could not complete - `page_count: optional number` Number of analysed pages (1 for images/docx) - `page_dimensions: optional array of object { height, page, width }` Rendered pixel size per page, so region bboxes can be scaled onto the page - `height: number` Rendered page height in pixels - `page: number` 0-based page index (0 for standalone images) - `width: number` Rendered page width in pixels - `reasoning: optional string` Explanation of the verdict - `suspect_regions: optional array of object { bbox, explanation, kind, 5 more }` Regions that led to the suspected fraud, ranked most-suspect first, each with an explanation of what makes it suspect - `bbox: array of number` Region bounding box as [x, y, w, h] in page-render pixels - `explanation: string` Human-readable explanation of what makes this region suspect - `kind: string` Kind of anomaly detected in this region - `page: number` 0-based page index (0 for standalone images) - `score: number` Suspicion score for this region (0 to 1) - `source: string` Detector that flagged this region - `primary: optional boolean` Whether this region is part of the small set of decisive evidence behind the verdict — the boxes a reviewer should look at first - `review: optional string` Automated reviewer verdict for this region (confirmed, dismissed, unsure, or empty). A dismissed region can still be surfaced when it is the only place to look; this label says how to read it - `synthetic_score: optional number` Likelihood (0 to 1) that the document is wholly generated or fabricated rather than a capture of a real document. Null for jobs completed before this score was introduced - `tampering_score: optional number` Likelihood (0 to 1) that a real captured document was locally edited — a genuine capture with regions altered after the fact. Null for jobs completed before this score was introduced - `transaction_id: optional string` Idempotency key - `updated_at: optional string` Update datetime ### Example ```http curl https://api.cloud.llamaindex.ai/api/alpha/verify/$JOB_ID \ -H "Authorization: Bearer $LLAMA_CLOUD_API_KEY" ``` #### Response ```json { "id": "id", "configuration": { "target_pages": "1,3,5-7", "tier": "agentic" }, "document_input_type": "file_id", "file_input": "file_input", "project_id": "project_id", "status": "CANCELLED", "user_id": "user_id", "created_at": "2019-12-27T18:11:19.117Z", "error_message": "error_message", "result": { "detector_version": "detector_version", "overall_score": 0, "verdict": "AUTHENTIC", "composite_scores": { "ai_generated": { "applicable": true, "score": 0 }, "document_coherence": { "applicable": true, "score": 0 }, "document_metadata": { "applicable": true, "score": 0 }, "known_fraud": { "applicable": true, "score": 0 }, "manually_edited": { "applicable": true, "score": 0 }, "recapture": { "applicable": true, "score": 0 } }, "confidence": 0, "error": "error", "page_count": 1, "page_dimensions": [ { "height": 0, "page": 0, "width": 0 } ], "reasoning": "reasoning", "suspect_regions": [ { "bbox": [ 0, 0, 0, 0 ], "explanation": "explanation", "kind": "kind", "page": 0, "score": 0, "source": "source", "primary": true, "review": "review" } ], "synthetic_score": 0, "tampering_score": 0 }, "transaction_id": "transaction_id", "updated_at": "2019-12-27T18:11:19.117Z" } ``` ## Cancel Verify Job **post** `/api/alpha/verify/{job_id}/cancel` Cancel a running Verify job. Stops processing and marks the job as CANCELLED. Returns the updated job. Jobs already in a terminal state (COMPLETED, FAILED, CANCELLED) cannot be cancelled. ### Path Parameters - `job_id: string` ### Query Parameters - `organization_id: optional string` - `project_id: optional string` ### Cookie Parameters - `session: optional string` ### Returns - `id: string` Unique identifier - `configuration: object { target_pages, tier }` Verify configuration used for this job - `target_pages: optional string` Comma-separated page numbers or ranges to analyze (1-based). Omit to analyze all pages. Ignored for non-PDF inputs. - `tier: optional "agentic" or "fast"` Verify tier: 'fast' runs only the quick deterministic forensic checks (metadata, content integrity, container structure, pixel statistics); 'agentic' (default) runs the full pipeline including the learned detectors and the semantic review pass. - `"agentic"` - `"fast"` - `document_input_type: "file_id" or "parse_job_id" or "url"` Type of the document input (FILE) - `"file_id"` - `"parse_job_id"` - `"url"` - `file_input: string` ID of the input file - `project_id: string` Project this job belongs to - `status: "CANCELLED" or "COMPLETED" or "FAILED" or 2 more` Current job status: PENDING, RUNNING, COMPLETED, FAILED, or CANCELLED - `"CANCELLED"` - `"COMPLETED"` - `"FAILED"` - `"PENDING"` - `"RUNNING"` - `user_id: string` User who created this job - `created_at: optional string` Creation datetime - `error_message: optional string` Error message if job failed - `result: optional object { detector_version, overall_score, verdict, 9 more }` Result of a Verify (doctored-document) analysis. Raw per-signal detail (evidence list, per-family sub-scores, raw regions, forensic heatmaps) is available separately via the job's details endpoint. - `detector_version: string` Version of the detector that produced the result - `overall_score: number` Overall doctoring likelihood (0 to 1) - `verdict: "AUTHENTIC" or "DOCTORED" or "LIKELY_DOCTORED" or 2 more` Overall verdict for the document - `"AUTHENTIC"` - `"DOCTORED"` - `"LIKELY_DOCTORED"` - `"NO_STRONG_SIGNAL"` - `"SUSPICIOUS"` - `composite_scores: optional object { ai_generated, document_coherence, document_metadata, 3 more }` Composite scores, each answering one question about the document - `ai_generated: optional object { applicable, score }` Was this content synthesized by a generative model? - `applicable: optional boolean` Whether the checks feeding this composite ran on this document. When false the document was not checked for this — not cleared of it - `score: optional number` Score (0 to 1); null when the composite was not applicable - `document_coherence: optional object { applicable, score }` Does the document's content agree with itself (checksums, arithmetic, machine-readable zones)? - `applicable: optional boolean` Whether the checks feeding this composite ran on this document. When false the document was not checked for this — not cleared of it - `score: optional number` Score (0 to 1); null when the composite was not applicable - `document_metadata: optional object { applicable, score }` Does the file's provenance / toolchain history look suspicious? Advisory: individually weak workflow-hygiene signals - `applicable: optional boolean` Whether the checks feeding this composite ran on this document. When false the document was not checked for this — not cleared of it - `score: optional number` Score (0 to 1); null when the composite was not applicable - `known_fraud: optional object { applicable, score }` Has this asset (or its template) been seen in fraud before? - `applicable: optional boolean` Whether the checks feeding this composite ran on this document. When false the document was not checked for this — not cleared of it - `score: optional number` Score (0 to 1); null when the composite was not applicable - `manually_edited: optional object { applicable, score }` Was this document altered after creation (splice, retype, redact, inpaint)? - `applicable: optional boolean` Whether the checks feeding this composite ran on this document. When false the document was not checked for this — not cleared of it - `score: optional number` Score (0 to 1); null when the composite was not applicable - `recapture: optional object { applicable, score }` Was the document captured through a channel that destroys forensic evidence (photo of a screen, print-then-rescan)? - `applicable: optional boolean` Whether the checks feeding this composite ran on this document. When false the document was not checked for this — not cleared of it - `score: optional number` Score (0 to 1); null when the composite was not applicable - `confidence: optional number` Confidence in the verdict (0 to 1): how firmly the detected signals support the verdict bucket, independent of the doctoring likelihood itself - `error: optional string` Error detail when the analysis could not complete - `page_count: optional number` Number of analysed pages (1 for images/docx) - `page_dimensions: optional array of object { height, page, width }` Rendered pixel size per page, so region bboxes can be scaled onto the page - `height: number` Rendered page height in pixels - `page: number` 0-based page index (0 for standalone images) - `width: number` Rendered page width in pixels - `reasoning: optional string` Explanation of the verdict - `suspect_regions: optional array of object { bbox, explanation, kind, 5 more }` Regions that led to the suspected fraud, ranked most-suspect first, each with an explanation of what makes it suspect - `bbox: array of number` Region bounding box as [x, y, w, h] in page-render pixels - `explanation: string` Human-readable explanation of what makes this region suspect - `kind: string` Kind of anomaly detected in this region - `page: number` 0-based page index (0 for standalone images) - `score: number` Suspicion score for this region (0 to 1) - `source: string` Detector that flagged this region - `primary: optional boolean` Whether this region is part of the small set of decisive evidence behind the verdict — the boxes a reviewer should look at first - `review: optional string` Automated reviewer verdict for this region (confirmed, dismissed, unsure, or empty). A dismissed region can still be surfaced when it is the only place to look; this label says how to read it - `synthetic_score: optional number` Likelihood (0 to 1) that the document is wholly generated or fabricated rather than a capture of a real document. Null for jobs completed before this score was introduced - `tampering_score: optional number` Likelihood (0 to 1) that a real captured document was locally edited — a genuine capture with regions altered after the fact. Null for jobs completed before this score was introduced - `transaction_id: optional string` Idempotency key - `updated_at: optional string` Update datetime ### Example ```http curl https://api.cloud.llamaindex.ai/api/alpha/verify/$JOB_ID/cancel \ -X POST \ -H "Authorization: Bearer $LLAMA_CLOUD_API_KEY" ``` #### Response ```json { "id": "id", "configuration": { "target_pages": "1,3,5-7", "tier": "agentic" }, "document_input_type": "file_id", "file_input": "file_input", "project_id": "project_id", "status": "CANCELLED", "user_id": "user_id", "created_at": "2019-12-27T18:11:19.117Z", "error_message": "error_message", "result": { "detector_version": "detector_version", "overall_score": 0, "verdict": "AUTHENTIC", "composite_scores": { "ai_generated": { "applicable": true, "score": 0 }, "document_coherence": { "applicable": true, "score": 0 }, "document_metadata": { "applicable": true, "score": 0 }, "known_fraud": { "applicable": true, "score": 0 }, "manually_edited": { "applicable": true, "score": 0 }, "recapture": { "applicable": true, "score": 0 } }, "confidence": 0, "error": "error", "page_count": 1, "page_dimensions": [ { "height": 0, "page": 0, "width": 0 } ], "reasoning": "reasoning", "suspect_regions": [ { "bbox": [ 0, 0, 0, 0 ], "explanation": "explanation", "kind": "kind", "page": 0, "score": 0, "source": "source", "primary": true, "review": "review" } ], "synthetic_score": 0, "tampering_score": 0 }, "transaction_id": "transaction_id", "updated_at": "2019-12-27T18:11:19.117Z" } ``` ## Get Verify Job Details **get** `/api/alpha/verify/{job_id}/details` Get the raw per-signal detail for a completed Verify job. Forensic drill-down behind the simplified result: the full evidence list, per-family sub-scores, raw localized regions, and per-page forensic heatmap overlays (presigned image URLs). ### Path Parameters - `job_id: string` ### Query Parameters - `organization_id: optional string` - `project_id: optional string` ### Cookie Parameters - `session: optional string` ### Returns - `job_id: string` ID of the Verify job - `degraded_tools: optional array of object { tool, reason }` Checks that could not run on this job (with the reason). A check listed here produced no findings because it could not run, not because the document is clean - `tool: string` Name of the check - `reason: optional string` Why the check could not run - `evidence: optional array of object { code, detail, family, 4 more }` Evidence items produced by detection tools - `code: string` Machine-readable evidence code - `detail: string` Human-readable evidence detail - `family: string` Signal family (e.g. metadata, splicing, compression) - `score: number` Evidence strength score - `tool: string` Tool that produced this evidence - `hard: optional boolean` Whether this is hard (conclusive) evidence - `metadata_tool: optional map[unknown]` Raw values from the tool behind this finding; the shape varies by tool - `heatmaps: optional array of object { expires_at, kind, page, 2 more }` Per-page forensic heatmap overlays as presigned image URLs - `expires_at: string` The time at which the presigned URL expires - `kind: string` Producing signal, e.g. double_compression, ela, noise - `page: number` 0-based page index (0 for standalone images) - `url: string` Presigned URL to the heatmap PNG (page overlay) - `score: optional number` Producing tool's max score (for ranking) - `page_dimensions: optional array of object { height, page, width }` Rendered pixel size per page, so region bboxes can be scaled onto the page - `height: number` Rendered page height in pixels - `page: number` 0-based page index (0 for standalone images) - `width: number` Rendered page width in pixels - `regions: optional array of object { bbox, detail, kind, 6 more }` Suspicious regions localized on rendered pages - `bbox: array of number` Region bounding box as [x, y, w, h] in page-render pixels - `detail: string` Human-readable detail about the region - `kind: string` Kind of anomaly detected in this region - `page: number` 0-based page index (0 for standalone images) - `score: number` Region-level doctoring likelihood score - `source: string` Detector/tool that produced this region - `primary: optional boolean` Whether this region is part of the small set of decisive evidence behind the verdict — the boxes a reviewer should look at first - `review: optional string` Review status/verdict for this region - `review_note: optional string` Free-form review note for this region - `sub_scores: optional map[number]` Per-family scores (metadata, ai_generation, splicing, copy_move, compression, noise, coherence, pdf_structure) ### Example ```http curl https://api.cloud.llamaindex.ai/api/alpha/verify/$JOB_ID/details \ -H "Authorization: Bearer $LLAMA_CLOUD_API_KEY" ``` #### Response ```json { "job_id": "job_id", "degraded_tools": [ { "tool": "tool", "reason": "reason" } ], "evidence": [ { "code": "code", "detail": "detail", "family": "family", "score": 0, "tool": "tool", "hard": true, "metadata_tool": { "foo": "bar" } } ], "heatmaps": [ { "expires_at": "2019-12-27T18:11:19.117Z", "kind": "kind", "page": 0, "url": "url", "score": 0 } ], "page_dimensions": [ { "height": 0, "page": 0, "width": 0 } ], "regions": [ { "bbox": [ 0, 0, 0, 0 ], "detail": "detail", "kind": "kind", "page": 0, "score": 0, "source": "source", "primary": true, "review": "review", "review_note": "review_note" } ], "sub_scores": { "foo": 0 } } ``` ## Domain Types ### Verify List Response - `VerifyListResponse object { id, configuration, document_input_type, 9 more }` Response for a Verify job. - `id: string` Unique identifier - `configuration: object { target_pages, tier }` Verify configuration used for this job - `target_pages: optional string` Comma-separated page numbers or ranges to analyze (1-based). Omit to analyze all pages. Ignored for non-PDF inputs. - `tier: optional "agentic" or "fast"` Verify tier: 'fast' runs only the quick deterministic forensic checks (metadata, content integrity, container structure, pixel statistics); 'agentic' (default) runs the full pipeline including the learned detectors and the semantic review pass. - `"agentic"` - `"fast"` - `document_input_type: "file_id" or "parse_job_id" or "url"` Type of the document input (FILE) - `"file_id"` - `"parse_job_id"` - `"url"` - `file_input: string` ID of the input file - `project_id: string` Project this job belongs to - `status: "CANCELLED" or "COMPLETED" or "FAILED" or 2 more` Current job status: PENDING, RUNNING, COMPLETED, FAILED, or CANCELLED - `"CANCELLED"` - `"COMPLETED"` - `"FAILED"` - `"PENDING"` - `"RUNNING"` - `user_id: string` User who created this job - `created_at: optional string` Creation datetime - `error_message: optional string` Error message if job failed - `result: optional object { detector_version, overall_score, verdict, 9 more }` Result of a Verify (doctored-document) analysis. Raw per-signal detail (evidence list, per-family sub-scores, raw regions, forensic heatmaps) is available separately via the job's details endpoint. - `detector_version: string` Version of the detector that produced the result - `overall_score: number` Overall doctoring likelihood (0 to 1) - `verdict: "AUTHENTIC" or "DOCTORED" or "LIKELY_DOCTORED" or 2 more` Overall verdict for the document - `"AUTHENTIC"` - `"DOCTORED"` - `"LIKELY_DOCTORED"` - `"NO_STRONG_SIGNAL"` - `"SUSPICIOUS"` - `composite_scores: optional object { ai_generated, document_coherence, document_metadata, 3 more }` Composite scores, each answering one question about the document - `ai_generated: optional object { applicable, score }` Was this content synthesized by a generative model? - `applicable: optional boolean` Whether the checks feeding this composite ran on this document. When false the document was not checked for this — not cleared of it - `score: optional number` Score (0 to 1); null when the composite was not applicable - `document_coherence: optional object { applicable, score }` Does the document's content agree with itself (checksums, arithmetic, machine-readable zones)? - `applicable: optional boolean` Whether the checks feeding this composite ran on this document. When false the document was not checked for this — not cleared of it - `score: optional number` Score (0 to 1); null when the composite was not applicable - `document_metadata: optional object { applicable, score }` Does the file's provenance / toolchain history look suspicious? Advisory: individually weak workflow-hygiene signals - `applicable: optional boolean` Whether the checks feeding this composite ran on this document. When false the document was not checked for this — not cleared of it - `score: optional number` Score (0 to 1); null when the composite was not applicable - `known_fraud: optional object { applicable, score }` Has this asset (or its template) been seen in fraud before? - `applicable: optional boolean` Whether the checks feeding this composite ran on this document. When false the document was not checked for this — not cleared of it - `score: optional number` Score (0 to 1); null when the composite was not applicable - `manually_edited: optional object { applicable, score }` Was this document altered after creation (splice, retype, redact, inpaint)? - `applicable: optional boolean` Whether the checks feeding this composite ran on this document. When false the document was not checked for this — not cleared of it - `score: optional number` Score (0 to 1); null when the composite was not applicable - `recapture: optional object { applicable, score }` Was the document captured through a channel that destroys forensic evidence (photo of a screen, print-then-rescan)? - `applicable: optional boolean` Whether the checks feeding this composite ran on this document. When false the document was not checked for this — not cleared of it - `score: optional number` Score (0 to 1); null when the composite was not applicable - `confidence: optional number` Confidence in the verdict (0 to 1): how firmly the detected signals support the verdict bucket, independent of the doctoring likelihood itself - `error: optional string` Error detail when the analysis could not complete - `page_count: optional number` Number of analysed pages (1 for images/docx) - `page_dimensions: optional array of object { height, page, width }` Rendered pixel size per page, so region bboxes can be scaled onto the page - `height: number` Rendered page height in pixels - `page: number` 0-based page index (0 for standalone images) - `width: number` Rendered page width in pixels - `reasoning: optional string` Explanation of the verdict - `suspect_regions: optional array of object { bbox, explanation, kind, 5 more }` Regions that led to the suspected fraud, ranked most-suspect first, each with an explanation of what makes it suspect - `bbox: array of number` Region bounding box as [x, y, w, h] in page-render pixels - `explanation: string` Human-readable explanation of what makes this region suspect - `kind: string` Kind of anomaly detected in this region - `page: number` 0-based page index (0 for standalone images) - `score: number` Suspicion score for this region (0 to 1) - `source: string` Detector that flagged this region - `primary: optional boolean` Whether this region is part of the small set of decisive evidence behind the verdict — the boxes a reviewer should look at first - `review: optional string` Automated reviewer verdict for this region (confirmed, dismissed, unsure, or empty). A dismissed region can still be surfaced when it is the only place to look; this label says how to read it - `synthetic_score: optional number` Likelihood (0 to 1) that the document is wholly generated or fabricated rather than a capture of a real document. Null for jobs completed before this score was introduced - `tampering_score: optional number` Likelihood (0 to 1) that a real captured document was locally edited — a genuine capture with regions altered after the fact. Null for jobs completed before this score was introduced - `transaction_id: optional string` Idempotency key - `updated_at: optional string` Update datetime ### Verify Create Response - `VerifyCreateResponse object { id, configuration, document_input_type, 9 more }` Response for a Verify job. - `id: string` Unique identifier - `configuration: object { target_pages, tier }` Verify configuration used for this job - `target_pages: optional string` Comma-separated page numbers or ranges to analyze (1-based). Omit to analyze all pages. Ignored for non-PDF inputs. - `tier: optional "agentic" or "fast"` Verify tier: 'fast' runs only the quick deterministic forensic checks (metadata, content integrity, container structure, pixel statistics); 'agentic' (default) runs the full pipeline including the learned detectors and the semantic review pass. - `"agentic"` - `"fast"` - `document_input_type: "file_id" or "parse_job_id" or "url"` Type of the document input (FILE) - `"file_id"` - `"parse_job_id"` - `"url"` - `file_input: string` ID of the input file - `project_id: string` Project this job belongs to - `status: "CANCELLED" or "COMPLETED" or "FAILED" or 2 more` Current job status: PENDING, RUNNING, COMPLETED, FAILED, or CANCELLED - `"CANCELLED"` - `"COMPLETED"` - `"FAILED"` - `"PENDING"` - `"RUNNING"` - `user_id: string` User who created this job - `created_at: optional string` Creation datetime - `error_message: optional string` Error message if job failed - `result: optional object { detector_version, overall_score, verdict, 9 more }` Result of a Verify (doctored-document) analysis. Raw per-signal detail (evidence list, per-family sub-scores, raw regions, forensic heatmaps) is available separately via the job's details endpoint. - `detector_version: string` Version of the detector that produced the result - `overall_score: number` Overall doctoring likelihood (0 to 1) - `verdict: "AUTHENTIC" or "DOCTORED" or "LIKELY_DOCTORED" or 2 more` Overall verdict for the document - `"AUTHENTIC"` - `"DOCTORED"` - `"LIKELY_DOCTORED"` - `"NO_STRONG_SIGNAL"` - `"SUSPICIOUS"` - `composite_scores: optional object { ai_generated, document_coherence, document_metadata, 3 more }` Composite scores, each answering one question about the document - `ai_generated: optional object { applicable, score }` Was this content synthesized by a generative model? - `applicable: optional boolean` Whether the checks feeding this composite ran on this document. When false the document was not checked for this — not cleared of it - `score: optional number` Score (0 to 1); null when the composite was not applicable - `document_coherence: optional object { applicable, score }` Does the document's content agree with itself (checksums, arithmetic, machine-readable zones)? - `applicable: optional boolean` Whether the checks feeding this composite ran on this document. When false the document was not checked for this — not cleared of it - `score: optional number` Score (0 to 1); null when the composite was not applicable - `document_metadata: optional object { applicable, score }` Does the file's provenance / toolchain history look suspicious? Advisory: individually weak workflow-hygiene signals - `applicable: optional boolean` Whether the checks feeding this composite ran on this document. When false the document was not checked for this — not cleared of it - `score: optional number` Score (0 to 1); null when the composite was not applicable - `known_fraud: optional object { applicable, score }` Has this asset (or its template) been seen in fraud before? - `applicable: optional boolean` Whether the checks feeding this composite ran on this document. When false the document was not checked for this — not cleared of it - `score: optional number` Score (0 to 1); null when the composite was not applicable - `manually_edited: optional object { applicable, score }` Was this document altered after creation (splice, retype, redact, inpaint)? - `applicable: optional boolean` Whether the checks feeding this composite ran on this document. When false the document was not checked for this — not cleared of it - `score: optional number` Score (0 to 1); null when the composite was not applicable - `recapture: optional object { applicable, score }` Was the document captured through a channel that destroys forensic evidence (photo of a screen, print-then-rescan)? - `applicable: optional boolean` Whether the checks feeding this composite ran on this document. When false the document was not checked for this — not cleared of it - `score: optional number` Score (0 to 1); null when the composite was not applicable - `confidence: optional number` Confidence in the verdict (0 to 1): how firmly the detected signals support the verdict bucket, independent of the doctoring likelihood itself - `error: optional string` Error detail when the analysis could not complete - `page_count: optional number` Number of analysed pages (1 for images/docx) - `page_dimensions: optional array of object { height, page, width }` Rendered pixel size per page, so region bboxes can be scaled onto the page - `height: number` Rendered page height in pixels - `page: number` 0-based page index (0 for standalone images) - `width: number` Rendered page width in pixels - `reasoning: optional string` Explanation of the verdict - `suspect_regions: optional array of object { bbox, explanation, kind, 5 more }` Regions that led to the suspected fraud, ranked most-suspect first, each with an explanation of what makes it suspect - `bbox: array of number` Region bounding box as [x, y, w, h] in page-render pixels - `explanation: string` Human-readable explanation of what makes this region suspect - `kind: string` Kind of anomaly detected in this region - `page: number` 0-based page index (0 for standalone images) - `score: number` Suspicion score for this region (0 to 1) - `source: string` Detector that flagged this region - `primary: optional boolean` Whether this region is part of the small set of decisive evidence behind the verdict — the boxes a reviewer should look at first - `review: optional string` Automated reviewer verdict for this region (confirmed, dismissed, unsure, or empty). A dismissed region can still be surfaced when it is the only place to look; this label says how to read it - `synthetic_score: optional number` Likelihood (0 to 1) that the document is wholly generated or fabricated rather than a capture of a real document. Null for jobs completed before this score was introduced - `tampering_score: optional number` Likelihood (0 to 1) that a real captured document was locally edited — a genuine capture with regions altered after the fact. Null for jobs completed before this score was introduced - `transaction_id: optional string` Idempotency key - `updated_at: optional string` Update datetime ### Verify Get Response - `VerifyGetResponse object { id, configuration, document_input_type, 9 more }` Response for a Verify job. - `id: string` Unique identifier - `configuration: object { target_pages, tier }` Verify configuration used for this job - `target_pages: optional string` Comma-separated page numbers or ranges to analyze (1-based). Omit to analyze all pages. Ignored for non-PDF inputs. - `tier: optional "agentic" or "fast"` Verify tier: 'fast' runs only the quick deterministic forensic checks (metadata, content integrity, container structure, pixel statistics); 'agentic' (default) runs the full pipeline including the learned detectors and the semantic review pass. - `"agentic"` - `"fast"` - `document_input_type: "file_id" or "parse_job_id" or "url"` Type of the document input (FILE) - `"file_id"` - `"parse_job_id"` - `"url"` - `file_input: string` ID of the input file - `project_id: string` Project this job belongs to - `status: "CANCELLED" or "COMPLETED" or "FAILED" or 2 more` Current job status: PENDING, RUNNING, COMPLETED, FAILED, or CANCELLED - `"CANCELLED"` - `"COMPLETED"` - `"FAILED"` - `"PENDING"` - `"RUNNING"` - `user_id: string` User who created this job - `created_at: optional string` Creation datetime - `error_message: optional string` Error message if job failed - `result: optional object { detector_version, overall_score, verdict, 9 more }` Result of a Verify (doctored-document) analysis. Raw per-signal detail (evidence list, per-family sub-scores, raw regions, forensic heatmaps) is available separately via the job's details endpoint. - `detector_version: string` Version of the detector that produced the result - `overall_score: number` Overall doctoring likelihood (0 to 1) - `verdict: "AUTHENTIC" or "DOCTORED" or "LIKELY_DOCTORED" or 2 more` Overall verdict for the document - `"AUTHENTIC"` - `"DOCTORED"` - `"LIKELY_DOCTORED"` - `"NO_STRONG_SIGNAL"` - `"SUSPICIOUS"` - `composite_scores: optional object { ai_generated, document_coherence, document_metadata, 3 more }` Composite scores, each answering one question about the document - `ai_generated: optional object { applicable, score }` Was this content synthesized by a generative model? - `applicable: optional boolean` Whether the checks feeding this composite ran on this document. When false the document was not checked for this — not cleared of it - `score: optional number` Score (0 to 1); null when the composite was not applicable - `document_coherence: optional object { applicable, score }` Does the document's content agree with itself (checksums, arithmetic, machine-readable zones)? - `applicable: optional boolean` Whether the checks feeding this composite ran on this document. When false the document was not checked for this — not cleared of it - `score: optional number` Score (0 to 1); null when the composite was not applicable - `document_metadata: optional object { applicable, score }` Does the file's provenance / toolchain history look suspicious? Advisory: individually weak workflow-hygiene signals - `applicable: optional boolean` Whether the checks feeding this composite ran on this document. When false the document was not checked for this — not cleared of it - `score: optional number` Score (0 to 1); null when the composite was not applicable - `known_fraud: optional object { applicable, score }` Has this asset (or its template) been seen in fraud before? - `applicable: optional boolean` Whether the checks feeding this composite ran on this document. When false the document was not checked for this — not cleared of it - `score: optional number` Score (0 to 1); null when the composite was not applicable - `manually_edited: optional object { applicable, score }` Was this document altered after creation (splice, retype, redact, inpaint)? - `applicable: optional boolean` Whether the checks feeding this composite ran on this document. When false the document was not checked for this — not cleared of it - `score: optional number` Score (0 to 1); null when the composite was not applicable - `recapture: optional object { applicable, score }` Was the document captured through a channel that destroys forensic evidence (photo of a screen, print-then-rescan)? - `applicable: optional boolean` Whether the checks feeding this composite ran on this document. When false the document was not checked for this — not cleared of it - `score: optional number` Score (0 to 1); null when the composite was not applicable - `confidence: optional number` Confidence in the verdict (0 to 1): how firmly the detected signals support the verdict bucket, independent of the doctoring likelihood itself - `error: optional string` Error detail when the analysis could not complete - `page_count: optional number` Number of analysed pages (1 for images/docx) - `page_dimensions: optional array of object { height, page, width }` Rendered pixel size per page, so region bboxes can be scaled onto the page - `height: number` Rendered page height in pixels - `page: number` 0-based page index (0 for standalone images) - `width: number` Rendered page width in pixels - `reasoning: optional string` Explanation of the verdict - `suspect_regions: optional array of object { bbox, explanation, kind, 5 more }` Regions that led to the suspected fraud, ranked most-suspect first, each with an explanation of what makes it suspect - `bbox: array of number` Region bounding box as [x, y, w, h] in page-render pixels - `explanation: string` Human-readable explanation of what makes this region suspect - `kind: string` Kind of anomaly detected in this region - `page: number` 0-based page index (0 for standalone images) - `score: number` Suspicion score for this region (0 to 1) - `source: string` Detector that flagged this region - `primary: optional boolean` Whether this region is part of the small set of decisive evidence behind the verdict — the boxes a reviewer should look at first - `review: optional string` Automated reviewer verdict for this region (confirmed, dismissed, unsure, or empty). A dismissed region can still be surfaced when it is the only place to look; this label says how to read it - `synthetic_score: optional number` Likelihood (0 to 1) that the document is wholly generated or fabricated rather than a capture of a real document. Null for jobs completed before this score was introduced - `tampering_score: optional number` Likelihood (0 to 1) that a real captured document was locally edited — a genuine capture with regions altered after the fact. Null for jobs completed before this score was introduced - `transaction_id: optional string` Idempotency key - `updated_at: optional string` Update datetime ### Verify Cancel Response - `VerifyCancelResponse object { id, configuration, document_input_type, 9 more }` Response for a Verify job. - `id: string` Unique identifier - `configuration: object { target_pages, tier }` Verify configuration used for this job - `target_pages: optional string` Comma-separated page numbers or ranges to analyze (1-based). Omit to analyze all pages. Ignored for non-PDF inputs. - `tier: optional "agentic" or "fast"` Verify tier: 'fast' runs only the quick deterministic forensic checks (metadata, content integrity, container structure, pixel statistics); 'agentic' (default) runs the full pipeline including the learned detectors and the semantic review pass. - `"agentic"` - `"fast"` - `document_input_type: "file_id" or "parse_job_id" or "url"` Type of the document input (FILE) - `"file_id"` - `"parse_job_id"` - `"url"` - `file_input: string` ID of the input file - `project_id: string` Project this job belongs to - `status: "CANCELLED" or "COMPLETED" or "FAILED" or 2 more` Current job status: PENDING, RUNNING, COMPLETED, FAILED, or CANCELLED - `"CANCELLED"` - `"COMPLETED"` - `"FAILED"` - `"PENDING"` - `"RUNNING"` - `user_id: string` User who created this job - `created_at: optional string` Creation datetime - `error_message: optional string` Error message if job failed - `result: optional object { detector_version, overall_score, verdict, 9 more }` Result of a Verify (doctored-document) analysis. Raw per-signal detail (evidence list, per-family sub-scores, raw regions, forensic heatmaps) is available separately via the job's details endpoint. - `detector_version: string` Version of the detector that produced the result - `overall_score: number` Overall doctoring likelihood (0 to 1) - `verdict: "AUTHENTIC" or "DOCTORED" or "LIKELY_DOCTORED" or 2 more` Overall verdict for the document - `"AUTHENTIC"` - `"DOCTORED"` - `"LIKELY_DOCTORED"` - `"NO_STRONG_SIGNAL"` - `"SUSPICIOUS"` - `composite_scores: optional object { ai_generated, document_coherence, document_metadata, 3 more }` Composite scores, each answering one question about the document - `ai_generated: optional object { applicable, score }` Was this content synthesized by a generative model? - `applicable: optional boolean` Whether the checks feeding this composite ran on this document. When false the document was not checked for this — not cleared of it - `score: optional number` Score (0 to 1); null when the composite was not applicable - `document_coherence: optional object { applicable, score }` Does the document's content agree with itself (checksums, arithmetic, machine-readable zones)? - `applicable: optional boolean` Whether the checks feeding this composite ran on this document. When false the document was not checked for this — not cleared of it - `score: optional number` Score (0 to 1); null when the composite was not applicable - `document_metadata: optional object { applicable, score }` Does the file's provenance / toolchain history look suspicious? Advisory: individually weak workflow-hygiene signals - `applicable: optional boolean` Whether the checks feeding this composite ran on this document. When false the document was not checked for this — not cleared of it - `score: optional number` Score (0 to 1); null when the composite was not applicable - `known_fraud: optional object { applicable, score }` Has this asset (or its template) been seen in fraud before? - `applicable: optional boolean` Whether the checks feeding this composite ran on this document. When false the document was not checked for this — not cleared of it - `score: optional number` Score (0 to 1); null when the composite was not applicable - `manually_edited: optional object { applicable, score }` Was this document altered after creation (splice, retype, redact, inpaint)? - `applicable: optional boolean` Whether the checks feeding this composite ran on this document. When false the document was not checked for this — not cleared of it - `score: optional number` Score (0 to 1); null when the composite was not applicable - `recapture: optional object { applicable, score }` Was the document captured through a channel that destroys forensic evidence (photo of a screen, print-then-rescan)? - `applicable: optional boolean` Whether the checks feeding this composite ran on this document. When false the document was not checked for this — not cleared of it - `score: optional number` Score (0 to 1); null when the composite was not applicable - `confidence: optional number` Confidence in the verdict (0 to 1): how firmly the detected signals support the verdict bucket, independent of the doctoring likelihood itself - `error: optional string` Error detail when the analysis could not complete - `page_count: optional number` Number of analysed pages (1 for images/docx) - `page_dimensions: optional array of object { height, page, width }` Rendered pixel size per page, so region bboxes can be scaled onto the page - `height: number` Rendered page height in pixels - `page: number` 0-based page index (0 for standalone images) - `width: number` Rendered page width in pixels - `reasoning: optional string` Explanation of the verdict - `suspect_regions: optional array of object { bbox, explanation, kind, 5 more }` Regions that led to the suspected fraud, ranked most-suspect first, each with an explanation of what makes it suspect - `bbox: array of number` Region bounding box as [x, y, w, h] in page-render pixels - `explanation: string` Human-readable explanation of what makes this region suspect - `kind: string` Kind of anomaly detected in this region - `page: number` 0-based page index (0 for standalone images) - `score: number` Suspicion score for this region (0 to 1) - `source: string` Detector that flagged this region - `primary: optional boolean` Whether this region is part of the small set of decisive evidence behind the verdict — the boxes a reviewer should look at first - `review: optional string` Automated reviewer verdict for this region (confirmed, dismissed, unsure, or empty). A dismissed region can still be surfaced when it is the only place to look; this label says how to read it - `synthetic_score: optional number` Likelihood (0 to 1) that the document is wholly generated or fabricated rather than a capture of a real document. Null for jobs completed before this score was introduced - `tampering_score: optional number` Likelihood (0 to 1) that a real captured document was locally edited — a genuine capture with regions altered after the fact. Null for jobs completed before this score was introduced - `transaction_id: optional string` Idempotency key - `updated_at: optional string` Update datetime ### Verify Get Details Response - `VerifyGetDetailsResponse object { job_id, degraded_tools, evidence, 4 more }` Raw per-signal detail for a completed Verify job. Forensic drill-down behind the simplified result: the full evidence list, per-family sub-scores, raw localized regions, and heatmap overlays. - `job_id: string` ID of the Verify job - `degraded_tools: optional array of object { tool, reason }` Checks that could not run on this job (with the reason). A check listed here produced no findings because it could not run, not because the document is clean - `tool: string` Name of the check - `reason: optional string` Why the check could not run - `evidence: optional array of object { code, detail, family, 4 more }` Evidence items produced by detection tools - `code: string` Machine-readable evidence code - `detail: string` Human-readable evidence detail - `family: string` Signal family (e.g. metadata, splicing, compression) - `score: number` Evidence strength score - `tool: string` Tool that produced this evidence - `hard: optional boolean` Whether this is hard (conclusive) evidence - `metadata_tool: optional map[unknown]` Raw values from the tool behind this finding; the shape varies by tool - `heatmaps: optional array of object { expires_at, kind, page, 2 more }` Per-page forensic heatmap overlays as presigned image URLs - `expires_at: string` The time at which the presigned URL expires - `kind: string` Producing signal, e.g. double_compression, ela, noise - `page: number` 0-based page index (0 for standalone images) - `url: string` Presigned URL to the heatmap PNG (page overlay) - `score: optional number` Producing tool's max score (for ranking) - `page_dimensions: optional array of object { height, page, width }` Rendered pixel size per page, so region bboxes can be scaled onto the page - `height: number` Rendered page height in pixels - `page: number` 0-based page index (0 for standalone images) - `width: number` Rendered page width in pixels - `regions: optional array of object { bbox, detail, kind, 6 more }` Suspicious regions localized on rendered pages - `bbox: array of number` Region bounding box as [x, y, w, h] in page-render pixels - `detail: string` Human-readable detail about the region - `kind: string` Kind of anomaly detected in this region - `page: number` 0-based page index (0 for standalone images) - `score: number` Region-level doctoring likelihood score - `source: string` Detector/tool that produced this region - `primary: optional boolean` Whether this region is part of the small set of decisive evidence behind the verdict — the boxes a reviewer should look at first - `review: optional string` Review status/verdict for this region - `review_note: optional string` Free-form review note for this region - `sub_scores: optional map[number]` Per-family scores (metadata, ai_generation, splicing, copy_move, compression, noise, coherence, pdf_structure)