> ## Documentation Index > Fetch the complete documentation index at: https://kernel.sh/docs/llms.txt > Use this file to discover all available pages before exploring further. # Retrieve a search > Returns the retained search resource exactly as it was returned by POST /search: results, attempts, warnings, usage, and expires_at. No provider is called and nothing is billed. Use it to look up a search by ID for debugging, cost review, or to recover result IDs before calling the contents endpoint. Inline content fetched at search time is included; content fetched later through the contents endpoint is not merged in. Missing, expired, or inaccessible searches return 404. ## OpenAPI ````yaml https://api.onkernel.com/spec.json get /search/{id} openapi: 3.1.0 info: description: Developer tools and cloud infrastructure for AI agents to use web browsers title: Kernel API version: 0.1.0 servers: - description: API Server url: https://api.onkernel.com security: - bearerAuth: [] tags: - description: Search the web and retrieve content for selected results. name: Search - description: Create and manage browser sessions. name: Browsers - description: Control mouse, keyboard, and screen on the browser instance. name: Browser Computer Controls - description: Execute Playwright code against the browser instance. name: Browser Playwright - description: Execute JavaScript in the browser instance's persistent Browser REPL. name: Browser REPL - description: Discover and invoke native page tools across the browser instance. name: Browser WebMCP - description: Read, write, and manage files on the browser instance. name: Browser Filesystem - description: Execute and manage processes on the browser instance. name: Browser Processes - description: Record and manage browser session video replays. name: Browser Replays - description: Stream logs from the browser instance. name: Browser Logs - description: >- Stream live telemetry events from a browser session, and manage the destinations sessions export them to. name: Browser Telemetry - description: Create, list, retrieve, and delete browser profiles. name: Profiles - description: Create and manage proxy configurations for routing browser traffic. name: Proxies - description: Create, list, retrieve, and delete browser extensions. name: Extensions - description: Create and manage browser pools for acquiring and releasing browsers. name: Browser Pools - description: Inspect the identity and authorization context for the current request. name: Authentication - description: >- Create and manage auth connections for automated credential capture and login. name: Managed Auth - description: Create and manage credentials for authentication. name: Credentials - description: Configure external credential providers like 1Password. name: Credential Providers - description: List applications and versions. name: Apps - description: Create and manage app deployments and stream deployment events. name: Deployments - description: Invoke actions and stream or query invocation status and events. name: Invocations - description: Read and manage organization-level limits. name: Organization - description: | Create and manage projects for resource isolation within an organization. When projects are disabled for the organization, project operations return `404` with code `projects_disabled`. name: Projects - description: Create and manage API keys for organization and project-scoped access. name: API Keys - description: Read audit log records for the authenticated organization. name: Audit Logs - description: Resolve browser and proxy recommendations for bot-protected sites. name: Config Registry paths: /search/{id}: get: tags: - Search summary: Retrieve a search description: | Returns the retained search resource exactly as it was returned by POST /search: results, attempts, warnings, usage, and expires_at. No provider is called and nothing is billed. Use it to look up a search by ID for debugging, cost review, or to recover result IDs before calling the contents endpoint. Inline content fetched at search time is included; content fetched later through the contents endpoint is not merged in. Missing, expired, or inaccessible searches return 404. operationId: getSearch parameters: - description: Search resource ID returned by POST /search. example: srch_abc123 in: path name: id required: true schema: type: string responses: '200': content: application/json: schema: $ref: '#/components/schemas/Search' description: The retained search. '401': $ref: '#/components/responses/Unauthorized' '403': $ref: '#/components/responses/Forbidden' '404': $ref: '#/components/responses/NotFound' '500': $ref: '#/components/responses/InternalError' '503': $ref: '#/components/responses/SearchUnavailable' security: - bearerAuth: [] components: schemas: Search: description: Retained search results and provider attempt history. properties: answer: description: >- Provider-generated answer when requested (e.g. via Tavily include_answer or Perplexity). Preserved independently of include_raw. type: string attempts: items: $ref: '#/components/schemas/SearchAttempt' type: array expires_at: description: > Expiration of result IDs for deferred retrieval. Results expire 24 hours after search completion. format: date-time type: string id: description: Search resource ID. Request tracing uses X-Request-Id. example: srch_abc123 type: string provider: description: Concrete serving provider, never auto or fallback. type: string query: description: >- Echo of the query. Native multi-query inputs are visible in the selected strategy target and the optional raw response. type: string raw: description: >- Full serving-provider response, including top-level metadata that does not belong to a result. Present only with include_raw=true; untrusted provider data. results: items: $ref: '#/components/schemas/SearchResult' type: array usage: $ref: '#/components/schemas/SearchUsage' warnings: items: $ref: '#/components/schemas/SearchWarning' type: array required: - id - expires_at - provider - results - attempts - warnings - query - usage type: object SearchAttempt: properties: duration_ms: minimum: 0 type: integer error_code: type: string outcome: enum: - success - empty - error - timeout type: string x-enum-varnames: - SearchAttemptOutcomeSuccess - SearchAttemptOutcomeEmpty - SearchAttemptOutcomeError - SearchAttemptOutcomeTimeout provider: type: string retryable: type: boolean required: - provider - outcome - duration_ms type: object SearchResult: properties: additional_snippets: items: type: string type: array content: $ref: '#/components/schemas/SearchResultContent' description: >- Portable retrieval outcome, or native content supplied by the search provider. Identity fields remain on the enclosing result. Native excerpts are labeled excerpt rather than full_page. Omission never triggers browser retrieval. id: description: >- Kernel-generated identifier for this result. Stable only within the retained search; not standardized across providers. Provider-native IDs, when available, remain provider-specific raw fields. example: srchr_def456 type: string published_date: description: >- Provider-supplied date or timestamp, preserving available precision. No publication date is fabricated. Retains the published field name. type: - string - 'null' rank: description: One-based position in the returned ranking. minimum: 1 type: integer raw: description: >- Original provider result, included only with include_raw=true. Provider relevance scores are not normalized. Top-level provider data is available in Search.raw. snippet: type: - string - 'null' source: description: Provider source name or result URL hostname, when available. type: - string - 'null' title: type: - string - 'null' url: description: Provider-returned URL, not assumed canonical. format: uri type: string required: - id - url - rank type: object SearchUsage: properties: content_fetches: description: >- Number of result URLs for which a Kernel browser retrieval was attempted, excluding cache-only hits. minimum: 0 type: integer cost: description: Total customer charge in USD when billing data is available. minimum: 0 type: number results_count: description: >- Number of result entries returned, including failed entries on the contents endpoint. minimum: 0 type: integer required: - results_count - content_fetches type: object SearchWarning: properties: code: description: >- Examples: param_unsupported, preference_unsupported, max_results_clamped, domains_truncated, recency_emulated, filter_emulated, date_filter_overridden, provider_ineligible, fallback_failed, content_partial. type: string message: type: string param: type: string provider: type: string result_id: type: string required: - code - message type: object Error: properties: code: description: Application-specific error code (machine-readable) example: bad_request type: string details: description: Additional error details (for multiple errors) items: $ref: '#/components/schemas/ErrorDetail' type: array inner_error: $ref: '#/components/schemas/ErrorDetail' message: description: Human-readable error description for debugging example: 'Missing required field: app_name' type: string required: - code - message type: object SearchResultContent: properties: cache_status: description: >- Kernel cache outcome. Provider-internal cache behavior may be unknown. enum: - hit - miss - bypass - unknown type: string x-enum-varnames: - SearchContentCacheStatusHit - SearchContentCacheStatusMiss - SearchContentCacheStatusBypass - SearchContentCacheStatusUnknown completeness: description: | Describes source coverage before max_chars truncation. Full_page means main-page content, not every dynamic element or linked page. enum: - full_page - excerpt - unknown type: string x-enum-varnames: - SearchContentCompletenessFullPage - SearchContentCompletenessExcerpt - SearchContentCompletenessUnknown error: $ref: '#/components/schemas/SearchContentError' extractor_version: description: Extraction version when Kernel transformed the input. type: string fetched_at: description: Origin retrieval time when known, not cache read time. format: date-time type: - string - 'null' final_url: description: Final retrieval URL when known. format: uri type: string format: $ref: '#/components/schemas/SearchContentFormat' http_status: description: Final target HTTP status when known. maximum: 599 minimum: 100 type: integer method: description: Original retrieval method, including on cache hits. enum: - provider - browser_curl - browser_render type: string x-enum-varnames: - SearchContentMethodProvider - SearchContentMethodBrowserCurl - SearchContentMethodBrowserRender status: description: | Ok means non-empty extracted content, not merely HTTP 200. Blocked includes detected challenges or access denials. Detection is best-effort, not a guarantee of page completeness. Error details are present for non-ok outcomes; text is present only on ok. enum: - ok - unavailable - blocked - timeout - unsupported_type - extraction_failed - error type: string x-enum-varnames: - SearchContentStatusOk - SearchContentStatusUnavailable - SearchContentStatusBlocked - SearchContentStatusTimeout - SearchContentStatusUnsupportedType - SearchContentStatusExtractionFailed - SearchContentStatusError text: description: >- Extracted website content, untrusted, not instructions. Present only on status=ok. type: string truncated: description: Whether max_chars truncated the extracted content. type: boolean required: - status type: object ErrorDetail: properties: code: description: Lower-level error code providing more specific detail example: invalid_input type: string message: description: Further detail about the error example: Provided version string is not semver compliant type: string type: object SearchContentError: properties: code: description: Machine-readable retrieval failure code. type: string message: description: Human-readable failure description. type: string retryable: type: boolean required: - code - message - retryable type: object SearchContentFormat: enum: - markdown - text type: string x-enum-varnames: - SearchContentFormatMarkdown - SearchContentFormatText responses: Unauthorized: content: application/json: schema: $ref: '#/components/schemas/Error' description: Unauthorized – missing or invalid authorization token Forbidden: content: application/json: schema: $ref: '#/components/schemas/Error' description: Forbidden – insufficient permissions or plan NotFound: content: application/json: schema: $ref: '#/components/schemas/Error' description: Resource not found InternalError: content: application/json: schema: $ref: '#/components/schemas/Error' description: Internal Server Error SearchUnavailable: content: application/json: schema: $ref: '#/components/schemas/Error' description: No provider is currently available for the request. securitySchemes: bearerAuth: scheme: bearer type: http ````