> For clean Markdown of any page, append .md to the page URL. > For a complete documentation index, see https://hiddenlayer.ferndocs.com/llms.txt. > For AI client integration (Claude Code, Cursor, etc.), connect to the MCP server at https://hiddenlayer.ferndocs.com/_mcp/server. # Get scan results GET https://api.hiddenlayer.ai/scan/v3/results/{scan_id} Reference: https://hiddenlayer.ferndocs.com/api-reference/llm-proxy-api/results/retrieve ## Authentication - `Authorization` header (bearer token, required) — Bearer authentication of the form `Bearer `, where token is your auth token. ## Servers - `https://api.hiddenlayer.ai` (ProdUs, default) - `https://api.eu.hiddenlayer.ai` (ProdEu) ## Request ### Path parameters - `scan_id` (string, required) ## Response ### 200 Successful response - `version` (string, required) — scanner version - `scan_id` (string, required) — unique identifier for the scan - `start_time` (datetime, required) — time the scan started - `status` (enum, required) — status of the scan - Allowed values: `pending`, `running`, `done`, `failed`, `canceled` - `inventory` (ScanModelInventoryComboV3, required) - `summary` (ScanReportOfRecordV3Summary, required) — A scan report without any file results. - `file_count` (integer, required, deprecated) — number of files scanned; use `.summary.file_count` instead - `files_with_detections_count` (integer, required, deprecated) — number of files with detections found; use `.summary.files_with_detections_count` instead - `detection_count` (integer, required, deprecated) — number of detections found; use `.summary.detection_count` instead - `has_genealogy` (boolean, optional) — if there is model geneaology info available - `$schema_version` (string, optional) — version of the scan report schema format - `end_time` (datetime, optional) — time the scan ended - `intelligence` (ModelIntelligence, optional) — Intelligence metadata about a model including origin, licensing, and usage policies - `compliance` (ScanReportOfRecordV3Compliance, optional) - `file_results` (list of ScanFileResultV3, optional) - `severity` (enum, optional) — The highest severity of any detections on the scan, including "safe". Use `.summary.highest_severity` instead. - Allowed values: `critical`, `high`, `medium`, `low`, `unknown`, `safe` - `detection_categories` (list of string, optional, deprecated) — list of detection categories found; use `.summary.detection_categories` instead ## Errors ### 400 Bad Request Error The request failed due to a client error, with one or more of the following possible causes: 1. The request required a tenant_id field, which was missing. 2. The request was malformed syntactically or semantically. - `any` ### 404 Not Found Error The specified resource was not found. - `any` ## Types ### ScanModelInventoryComboV3 - `model_name` (string, required) — name of the model - `requested_scan_location` (string, required) — Location to be scanned - `model_id` (string, required) — Unique identifier for the model - `model_version_id` (string, required) — unique identifier for the model version - `model_version` (string, optional) — version of the model - `model_source` (string, optional) — source (provider) info - `file_location` (string, optional) — URL or path to the model files, if available - `provider_details` (ProviderDetails, optional) - `asset_region` (string, optional) — Region of discovered asset - `requesting_entity` (string, optional) — Entity that requested the scan - `request_source` (enum, optional) — Identifies the system that requested the scan - Allowed values: `Hybrid Upload`, `API Upload`, `Integration`, `UI Upload`, `AI Asset Discovery` - `origin` (string, optional) — Specifies the platform or service where the model originated before being scanned ### ScanReportOfRecordV3Summary A scan report without any file results. - `version` (string, required) — scanner version - `scan_id` (string, required) — unique identifier for the scan - `start_time` (datetime, required) — time the scan started - `status` (enum, required) — status of the scan - Allowed values: `pending`, `running`, `done`, `failed`, `canceled` - `inventory` (ScanModelInventoryComboV3, required) - `summary` (ScanReportOfRecordV3SummarySummary, required) - `file_count` (integer, required, deprecated) — number of files scanned; use `.summary.file_count` instead - `files_with_detections_count` (integer, required, deprecated) — number of files with detections found; use `.summary.files_with_detections_count` instead - `detection_count` (integer, required, deprecated) — number of detections found; use `.summary.detection_count` instead - `has_genealogy` (boolean, optional) — if there is model geneaology info available - `$schema_version` (string, optional) — version of the scan report schema format - `end_time` (datetime, optional) — time the scan ended - `compliance` (ScanReportOfRecordV3SummaryCompliance, optional) - `severity` (enum, optional) — The highest severity of any detections on the scan, including "safe". Use `.summary.highest_severity` instead. - Allowed values: `critical`, `high`, `medium`, `low`, `unknown`, `safe` - `detection_categories` (list of string, optional, deprecated) — list of detection categories found; use `.summary.detection_categories` instead ### ModelIntelligence Intelligence metadata about a model including origin, licensing, and usage policies - `geographic_footprint` (list of string, optional) — List of countries where the model originated - `country_of_origin` (string, optional) — ISO 3166-1 alpha-2 country code of the model's primary origin - `contributor_trust_level` (string, optional) — Trust level of the model contributor - `licenses` (list of ModelLicense, optional) — List of licenses associated with the model - `usage_policies` (list of ModelUsagePolicy, optional) — List of usage policies associated with the model ### ScanReportOfRecordV3Compliance - `status` (enum, optional) - Allowed values: `COMPLIANT`, `NONCOMPLIANT` - `evaluated_at` (datetime, optional) — The datetime when the rule set was evaluated against the scan result - `rule_set_ids` (list of string, optional) — A list of non-default rule sets that were used when evaluating the scan result ### ScanFileResultV3 - `file_instance_id` (string, required) — unique ID of the file - `file_location` (string, required) — full file path - `start_time` (datetime, required) — time the scan started - `end_time` (datetime, required) — time the scan ended - `details` (FileDetailsV3, required) - `status` (enum, required) — status of the scan - Allowed values: `skipped`, `pending`, `running`, `done`, `failed`, `canceled` - `seen` (datetime, required) — time the scan was seen at - `detections` (list of ScanDetectionV3, required) - `advisories` (list of ScanAdvisoryV3, optional) — informational advisories associated with this file (e.g. tokenizer family) - `file_error` (list of string, optional) — Error messages returned by the scanner ### ProviderDetails - `provider` (enum, required) - Allowed values: `AWS_BEDROCK`, `AWS_SAGEMAKER`, `AZURE_AI_FOUNDRY`, `AZURE_ML`, `DATABRICKS` - `provider_model_id` (string, required) — The provider's unique identifier for the model. Examples: - AWS Bedrock: "anthropic.claude-3-5-sonnet-20241022-v2:0" - Azure AI Foundry: "Claude-3-5-Sonnet" - `model_arn` (string, optional) — Optional full ARN or resource identifier for the model. Used for provisioned models, custom deployments, or cross-account access. - `country` (string, optional) — Optional country code (ISO 3166-1 alpha-2) for the location where the model provider is primarily based. ### ScanReportOfRecordV3SummarySummary - `severity` (enum, optional) — The highest severity of any detections on the scan, including "safe". Use `.summary.highest_severity` instead. - Allowed values: `critical`, `high`, `medium`, `low`, `unknown`, `safe` - `highest_severity` (enum, optional) — The highest severity of any detections on the scan. - Allowed values: `critical`, `high`, `medium`, `low`, `none`, `unknown` - `detection_count` (integer, optional) — total number of detections found - `advisory_count` (integer, optional) — total number of advisories found - `file_count` (integer, optional) — total number of files scanned - `files_with_detections_count` (integer, optional) — number of files that contain detections - `detection_categories` (list of string, optional) — list of unique detection categories found - `advisory_categories` (list of string, optional) — list of unique advisory categories found - `files_failed_to_scan` (integer, optional) — number of files that failed during scanning - `unknown_files` (integer, optional) — number of files with unknown file type ### ScanReportOfRecordV3SummaryCompliance - `status` (enum, optional) - Allowed values: `COMPLIANT`, `NONCOMPLIANT` - `evaluated_at` (datetime, optional) — The datetime when the rule set was evaluated against the scan result - `rule_set_ids` (list of string, optional) — A list of non-default rule sets that were used when evaluating the scan result ### ModelLicense License information for a model - `name` (string, required) — Name of the license - `sha256` (string, required) — SHA256 hash of the license file ### ModelUsagePolicy Usage policy information for a model - `name` (string, required) — Name of the usage policy - `sha256` (string, required) — SHA256 hash of the policy document ### FileDetailsV3 - `estimated_time` (string, required) — estimated time to scan the file - `sha256` (string, required) — hexadecimal sha256 hash of file - `file_type` (string, required) — type of the file - `md5` (string, optional) — hexadecimal md5 hash of file - `tlsh` (string, optional) — TLSH hash of file - `file_size` (string, optional) — size of the file in human readable format - `file_size_bytes` (integer, optional) — size of the file in bytes - `file_type_details` (FileDetailsV3FileTypeDetails, optional) ### ScanDetectionV3 - `detection_id` (string, required) — unique identifier for the detection - `rule_id` (string, required) — unique identifier for the rule that sourced the detection - `risk` (enum, required) — detection risk - Allowed values: `MALICIOUS`, `SUSPICIOUS` - `category` (string, required) — Vulnerability category for the detection - `description` (string, required) — detection description - `likelihood` (string, required) — detection likelihood - `impact` (string, required) — detection impact - `severity` (enum, required) — The severity of the detection. - Allowed values: `critical`, `high`, `medium`, `low` - `mitre_atlas` (list of ScanDetectionV3MitreAtlasItems, required) - `owasp` (list of string, required) - `cve` (list of string, required) - `cwe` (string, required) - `cwe_href` (string, required) — CWE URL for the detection - `rule_details` (list of ScanDetectionV3RuleDetailsItems, optional) - `technical_blog_hrefs` (list of string, optional) — Hiddenlayer Technical Blog URLs for the detection - `technical_blog_href` (string, optional, deprecated) — Hiddenlayer Technical Blog URL for the detection ### ScanAdvisoryV3 An informational advisory associated with a file. Advisories carry guidance about a property of the model (e.g. tokenizer family) that may matter to a downstream consumer, but do not represent a concrete detection. - `advisory_id` (string, required) — unique identifier for the advisory - `rule_id` (string, required) — unique identifier for the rule that sourced the advisory - `category` (string, required) — category for the advisory - `description` (string, required) — advisory description ### FileDetailsV3FileTypeDetails ### ScanDetectionV3MitreAtlasItems - `technique` (string, optional) — MITRE Atlas Technique - `tactic` (string, optional) — MITRE Atlas Tactic ### ScanDetectionV3RuleDetailsItems - `status` (enum, optional) — status - Allowed values: `created`, `deprecated`, `updated`, `superseded` - `status_at` (datetime, optional) — date-time when the details entry was created - `description` (string, optional) — description of the deprecation ### GGUFFileAttributes - `subtype` (list of string, required) ### KerasFileAttributes - `subtype` (list of string, required) - `pickle_modules` (list of string, required) - `keras_version` (string, optional) — version of the Keras file - `keras_class_name` (string, optional) - `keras_date_saved_at` (string, optional) - `keras_module` (string, optional) ### NumpyFileAttributes - `subtype` (list of string, required) - `numpy_arrays` (string, required) - `numpy_shape` (list of string, required) ### RDSFileAttributes - `subtype` (list of string, required) - `rds_encoding` (string, required) — encoding of the RDS file - `rds_min_reader_version` (string, required) — minimum reader version for the RDS file - `rds_version` (string, required) — version of the RDS file - `rds_writer_version` (string, required) — version of the RDS writer ## Examples **Response** ```json { "version": "string", "scan_id": "string", "start_time": "2024-01-15T09:30:00Z", "status": "pending", "inventory": { "model_name": "keras-tf-2025-05-27", "requested_scan_location": "/files-to-scan", "model_id": "00000000-0000-0000-0000-000000000000", "model_version_id": "00000000-0000-0000-0000-000000000000", "model_version": "1.0.0", "model_source": "adhoc", "file_location": "https://huggingface.co/meta-llama/Llama-3.1-8B", "provider_details": { "provider": "AWS_BEDROCK", "provider_model_id": "anthropic.claude-3-5-sonnet-20241022-v2:0" }, "asset_region": "string", "requesting_entity": "string", "request_source": "Hybrid Upload", "origin": "Hugging Face" }, "summary": { "file_count": 1, "files_with_detections_count": 1, "detection_count": 1, "severity": "critical", "detection_categories": [ "string" ], "highest_severity": "critical", "advisory_count": 1, "advisory_categories": [ "string" ], "files_failed_to_scan": 1, "unknown_files": 1 }, "file_count": 1, "files_with_detections_count": 1, "detection_count": 1, "has_genealogy": true, "$schema_version": "string", "end_time": "2024-01-15T09:30:00Z", "intelligence": { "geographic_footprint": [ "US", "GB" ], "country_of_origin": "US", "contributor_trust_level": "high", "licenses": [ { "name": "Apache-2.0", "sha256": "abc123..." } ], "usage_policies": [ { "name": "Commercial Use Allowed", "sha256": "def456..." } ] }, "compliance": { "status": "COMPLIANT", "evaluated_at": "2024-01-15T09:30:00Z", "rule_set_ids": [ "string" ] }, "file_results": [ { "file_instance_id": "string", "file_location": "string", "start_time": "2024-10-16T23:38:32.278Z", "end_time": "2024-10-16T23:38:32.354Z", "details": { "estimated_time": "string", "sha256": "a54d88e06612d820bc3be72877c74f257b561b19", "file_type": "safetensors", "md5": "ce114e4501d2f4e2dcea3e17b546f339", "tlsh": "T1C50757F93C74D00C05B70C0793A1D5A9DF3F6D3A2F7AD940F3BFBF07B3BDF5A1D293", "file_size": "9 GB", "file_size_bytes": 9663676416, "file_type_details": { "subtype": [ "string" ] } }, "status": "skipped", "seen": "2024-10-22T17:59:12.431Z", "detections": [ { "detection_id": "00000000-0000-0000-0000-000000000000", "rule_id": "PICKLE_0055_202408", "risk": "MALICIOUS", "category": "Arbitrary Code Execution", "description": "Found lambda embedded in keras model allowing custom layers that support arbitrary expression execution", "likelihood": "medium", "impact": "critical", "severity": "critical", "mitre_atlas": [ { "technique": "string", "tactic": "string" } ], "owasp": [ "string" ], "cve": [ "string" ], "cwe": "string", "cwe_href": "string", "rule_details": [ { "status": "created", "status_at": "2024-01-15T09:30:00Z", "description": "string" } ], "technical_blog_hrefs": [ "string" ], "technical_blog_href": "string" } ], "advisories": [ { "advisory_id": "00000000-0000-0000-0000-000000000000", "rule_id": "SAFETENSORS_0001_202512", "category": "TokenBreak", "description": "Models using the BPE and WordPiece tokenization strategies are vulnerable to TokenBreak" } ], "file_error": [ "File not found" ] } ], "severity": "critical", "detection_categories": [ "string" ] } ``` **SDK Code** ```python import requests url = "https://api.hiddenlayer.ai/scan/v3/results/scan_id" headers = {"Authorization": "Bearer "} response = requests.get(url, headers=headers) print(response.json()) ``` ```javascript const url = 'https://api.hiddenlayer.ai/scan/v3/results/scan_id'; const options = {method: 'GET', headers: {Authorization: 'Bearer '}}; try { const response = await fetch(url, options); const data = await response.json(); console.log(data); } catch (error) { console.error(error); } ``` ```go package main import ( "fmt" "net/http" "io" ) func main() { url := "https://api.hiddenlayer.ai/scan/v3/results/scan_id" req, _ := http.NewRequest("GET", url, nil) req.Header.Add("Authorization", "Bearer ") res, _ := http.DefaultClient.Do(req) defer res.Body.Close() body, _ := io.ReadAll(res.Body) fmt.Println(res) fmt.Println(string(body)) } ``` ```ruby require 'uri' require 'net/http' url = URI("https://api.hiddenlayer.ai/scan/v3/results/scan_id") http = Net::HTTP.new(url.host, url.port) http.use_ssl = true request = Net::HTTP::Get.new(url) request["Authorization"] = 'Bearer ' response = http.request(request) puts response.read_body ``` ```java import com.mashape.unirest.http.HttpResponse; import com.mashape.unirest.http.Unirest; HttpResponse response = Unirest.get("https://api.hiddenlayer.ai/scan/v3/results/scan_id") .header("Authorization", "Bearer ") .asString(); ``` ```php request('GET', 'https://api.hiddenlayer.ai/scan/v3/results/scan_id', [ 'headers' => [ 'Authorization' => 'Bearer ', ], ]); echo $response->getBody(); ``` ```csharp using RestSharp; var client = new RestClient("https://api.hiddenlayer.ai/scan/v3/results/scan_id"); var request = new RestRequest(Method.GET); request.AddHeader("Authorization", "Bearer "); IRestResponse response = client.Execute(request); ``` ```swift import Foundation let headers = ["Authorization": "Bearer "] let request = NSMutableURLRequest(url: NSURL(string: "https://api.hiddenlayer.ai/scan/v3/results/scan_id")! as URL, cachePolicy: .useProtocolCachePolicy, timeoutInterval: 10.0) request.httpMethod = "GET" request.allHTTPHeaderFields = headers let session = URLSession.shared let dataTask = session.dataTask(with: request as URLRequest, completionHandler: { (data, response, error) -> Void in if (error != nil) { print(error as Any) } else { let httpResponse = response as? HTTPURLResponse print(httpResponse) } }) dataTask.resume() ```