> ## Documentation Index > Fetch the complete documentation index at: https://docs.preuve.ai/llms.txt > Use this file to discover all available pages before exploring further. # Start a batch > Create up to 5 analysis runs in one batch. Idempotent on the batch clientRunId. Each item needs its own unique clientRunId. Item scanType is optional and falls back to the batch's, so set it per item only to mix starter and deep in one batch. ## OpenAPI ````yaml /api-reference/openapi.yaml post /api/agent/analysis-batches openapi: 3.1.0 info: title: Preuve Agent API version: '2.0' summary: Startup idea validation for AI agents and scripts. description: > API for programmatic startup idea validation: a 0-100 viability score, competitors, market sizing and source-linked evidence from 60+ live sources. Every request carries a single x-preuve-key header holding the whole API key - no signing, no bearer token. See the Authentication guide. Every response is JSON, errors included: an unknown route answers 404 with the same { error, code } envelope as every other failure, so branch on code and never parse a message. Error codes are stable and retired ones stay listed on the Errors page. scanType "starter" costs nothing and returns a real report, which makes it the sandbox: an agent can exercise the whole create - poll - enrich - export loop without spending a cent. GET /api/agent is the unauthenticated discovery document. termsOfService: https://preuve.ai/terms contact: name: Preuve AI developer support url: https://preuve.ai/support email: hello@preuve.ai servers: - url: https://preuve.ai security: - preuveKey: [] tags: - name: Discovery description: Unauthenticated documents that describe the API itself. - name: Analyses description: One idea at a time - create, poll, enrich, export. - name: Batches description: Up to five analyses in one idempotent batch. - name: Tools description: Utilities that spend no scan quota. - name: Agency description: >- Client work on a Consultant or Agency workspace (agency:read / agency:write). externalDocs: description: Developer docs - quickstart, authentication, quotas, error codes url: https://docs.preuve.ai/ paths: /api/agent/analysis-batches: post: tags: - Batches summary: Start a batch description: > Create up to 5 analysis runs in one batch. Idempotent on the batch clientRunId. Each item needs its own unique clientRunId. Item scanType is optional and falls back to the batch's, so set it per item only to mix starter and deep in one batch. operationId: createBatch requestBody: required: true content: application/json: schema: $ref: '#/components/schemas/CreateBatchRequest' responses: '200': description: > Existing batch replayed via clientRunId. Nothing new was created and nothing new was consumed; the body is the batch's current state. content: application/json: schema: $ref: '#/components/schemas/Batch' '202': description: > Batch accepted. Every item has been attempted, so a 202 can still carry items that failed at start - read status and counts, and each item's own status, rather than assuming five runs began. content: application/json: schema: $ref: '#/components/schemas/Batch' '400': $ref: '#/components/responses/BadRequest' '401': $ref: '#/components/responses/Unauthorized' '403': $ref: '#/components/responses/InsufficientScope' '409': $ref: '#/components/responses/ClientRunIdConflict' '413': $ref: '#/components/responses/PayloadTooLarge' '429': $ref: '#/components/responses/RateLimited' '500': $ref: '#/components/responses/InternalError' '503': $ref: '#/components/responses/ServiceUnavailable' components: schemas: CreateBatchRequest: type: object required: - clientRunId - items anyOf: - title: scanType set for the whole batch required: - scanType - title: scanType set on every item properties: items: items: required: - scanType properties: clientRunId: type: string description: Idempotency id for the whole batch. name: type: string scanType: type: string enum: - starter - deep description: Default scanType applied to every item; each item may override. enrichmentMode: type: string enum: - core - none items: type: array minItems: 1 maxItems: 5 items: $ref: '#/components/schemas/BatchAnalysisItem' Batch: type: object properties: id: type: string format: uuid clientRunId: type: string name: type: - string - 'null' status: type: string counts: type: object additionalProperties: type: integer items: type: array items: $ref: '#/components/schemas/BatchItem' BatchAnalysisItem: description: > One item in a batch. Same fields as CreateAnalysisRequest, except that scanType is optional: an item without one inherits the batch's scanType, and only a batch that sets neither is rejected. clientRunId is still required per item - the derived fallback the batch uses for its own reservation bookkeeping does not reach the run. allOf: - $ref: '#/components/schemas/AnalysisRequestFields' required: - clientRunId - idea BatchItem: description: > One entry in a batch. Either a real run or a late-collision stub; the two are told apart by id, which is null only on the stub. oneOf: - $ref: '#/components/schemas/AnalysisRun' - $ref: '#/components/schemas/BatchItemConflict' ApiError: type: object required: - error - code description: > Every error, on every route, unknown paths included. Some codes add top-level context fields (limit, tokenBalance, resetAt, upsell, ...); a 404 NOT_FOUND on an unknown route adds details.discovery, details.openapi and details.docs so the caller can recover. properties: error: type: string description: Human-readable message. Prose; it changes. code: type: string description: Stable machine-readable code. Branch on this. details: description: > Optional context. An object on most codes (limits, recovery links, the method and path of a 404); an array of strings where the route forwards a validator's messages, as enrich does on INVALID_FOUNDER_PROFILE. Read it, never branch on it - `code` is the contract. oneOf: - type: object additionalProperties: true - type: array items: type: string limit: type: integer description: > The ceiling that was hit, in the unit the code implies: bytes on PAYLOAD_TOO_LARGE, runs in flight on CONCURRENT_LIMIT_REACHED, requests in the window on a rate limit. remaining: type: integer description: Requests left in the current window. resetAt: type: string format: date-time description: > When the window that refused this request reopens. Present on DAILY_LIMIT_REACHED from run creation (the idea generator's rolling ceiling sends only limit); prefer it over the Retry-After header when both are present, since it is absolute. tokenBalance: type: integer description: >- Deep-scan credits left on the account. Present on INSUFFICIENT_TOKENS. scansUsed: type: integer scansRemaining: type: integer limitType: type: string description: Which ceiling refused an AppSumo-backed account. upgradeHint: type: string description: What to change to get past this refusal. Prose, not a contract. paused: type: boolean description: > Present only on the LLM-pause 503, where the body is exactly { error, code, paused }. upsell: $ref: '#/components/schemas/ExportUpsell' AnalysisRequestFields: type: object description: > Every field an analysis request can carry. Nothing is required here - CreateAnalysisRequest and BatchAnalysisItem each add their own required list, because a batch item may inherit scanType from the batch while a standalone create must state it. properties: clientRunId: type: string description: Caller-chosen idempotency id. Reusing it replays the existing run. scanType: type: string enum: - starter - deep description: >- "starter" costs nothing; "deep" runs the full paid analysis and consumes account quota. "free" is accepted as a legacy alias of "starter". idea: type: string description: The startup idea, in plain language. targetMarket: type: string targetCountry: type: string description: ISO country code, e.g. "US" or "FR". title: type: string stage: type: string description: >- Where the founder is today. Canonical values: idea, validation, mvp, launched, growth, scaling (free text tolerated). Optional here, but the MCP tools require it for deep scans; it feeds the feasibility analysis. budget: type: string description: >- Budget available to pursue the idea. Canonical values: bootstrap (no outside money), 10k, 50k, 100k, 100kPlus (over $100k), 1mPlus (over $1M); free text tolerated. Optional here, but the MCP tools require it for deep scans; it feeds the feasibility analysis. language: type: string default: en enrichmentMode: type: string enum: - core - none default: core description: >- "core" also generates the export-required sections; "none" returns the raw completion only. publish: type: boolean description: When true, creates a public share URL for the report. AnalysisRun: type: object properties: id: type: string format: uuid reportId: type: - string - 'null' format: uuid description: > Null until the report row is attached, which is why a run that failed before dispatch is retryable under the same clientRunId. clientRunId: type: - string - 'null' description: > The caller-chosen id from create. Null when the analysis is a dashboard-created report resolved by its report id: no run row exists, so there is no clientRunId to replay. batchId: type: - string - 'null' format: uuid description: The batch this run belongs to, or null for a standalone run. scanType: type: string enum: - starter - deep description: Your request type. Reliable from creation, even while PROCESSING. enrichmentMode: type: string enum: - core - none publish: type: boolean reportType: type: - string - 'null' enum: - quick - deep_dive - null analysisTier: type: - string - 'null' enum: - basic - advanced - null description: >- Derived from the report. Deep runs read "basic" until deep sections land. status: type: string enum: - PENDING - PROCESSING - COMPLETED - FAILED progressStep: type: - string - 'null' description: Free-text pipeline stage while PROCESSING. errorCode: type: - string - 'null' description: > Set on a FAILED run. Two internal markers are deliberately rewritten before they leave: an ambiguous dispatch reports TRIGGER_DISPATCH_FAILED and an ambiguous claim refund reports REPORT_CREATE_FAILED. errorMessage: type: - string - 'null' resetAt: type: string format: date-time description: > Present only on a batch item refused by the rolling daily cap; absent, not null, on every other run. pollAfterSeconds: type: integer description: > Present only while polling is still useful. Pace by this rather than by a fixed interval. pollHint: type: string description: Present alongside pollAfterSeconds. createdAt: type: string format: date-time updatedAt: type: string format: date-time readyForExport: type: boolean description: > Whether GET /export will serve this analysis now. Always false for a report created on the preuve.ai dashboard, whatever state the report is in, because that export answers 403 REPORT_NOT_FROM_AGENT_RUN. Branch on exportable first. exportable: type: boolean description: > Present, and false, only on a report created on the preuve.ai dashboard. Absent means the ordinary contract applies and readyForExport is the gate. When it is false, no amount of polling or enriching changes it. exportBlockedCode: type: string enum: - REPORT_NOT_FROM_AGENT_RUN description: Present alongside exportable, naming the refusal. exportBlockedHint: type: string description: Present alongside exportable. Relayable prose. enrichment: $ref: '#/components/schemas/EnrichmentState' modules: type: - object - 'null' description: Per-module statuses. Deep reports only; null on free runs. properties: proofOfDemand: $ref: '#/components/schemas/ModuleStatus' founderFit: $ref: '#/components/schemas/ModuleStatus' playbook: $ref: '#/components/schemas/ModuleStatus' trends: $ref: '#/components/schemas/ModuleStatus' community: $ref: '#/components/schemas/ModuleStatus' reportUrl: type: - string - 'null' format: uri description: Null until the report row exists. shareUrl: type: - string - 'null' format: uri BatchItemConflict: type: object description: > Not a run. Another request claimed this item's clientRunId after the batch row was reserved but before the item started, so nothing was created and id is null. Appears in Batch.items on create, on replay and on poll, and its null id carries through to BatchExport.omittedItems. properties: id: type: 'null' reportId: type: 'null' clientRunId: type: string batchId: type: string format: uuid status: type: string enum: - FAILED errorCode: type: string enum: - CLIENT_RUN_ID_CONFLICT errorMessage: type: string ExportUpsell: type: object description: > Purchase context for a locked export or buyable quota refusal. Export upsells are free-owner only; a paid non-regional caller may receive a scan_pack top-up on INSUFFICIENT_TOKENS. Ignore it if you only need the data contract. properties: offer: type: string enum: - founder_report - scan_pack description: > "founder_report" is the $59 one-time tier. "scan_pack" is a top-up for an account that already bought and has no deep credits left. price: type: string description: Display price text, e.g. "$59 one-time". guarantee: type: string description: > Guarantee text for offer=founder_report. Omitted for scan_pack top-ups; an empty string is never used. url: type: string format: uri description: Purchase or report URL, with attribution query parameters. locked: type: array description: > Section names the offer covers. Distinct from the top-level locked object, which names response FIELDS. items: type: string note: type: string description: The same facts as one sentence, for a client that relays text. EnrichmentState: type: object description: > Progress of the core export sections. The four name arrays partition `required`, so a poller decides what to do next by reading them rather than by parsing `status`: anything in processing means keep polling, anything in failed means call enrich again, and missing with nothing processing means enrichment was never started for it. properties: mode: type: string enum: - core - none status: type: string enum: - not_required - pending - processing - partial - completed - failed description: > Rolled up from the arrays below. "partial" means some sections landed and others did not; "not_required" is mode "none", or a report with no required sections. required: type: array description: >- Every section this report and mode need. The other four arrays partition it. items: type: string completed: type: array items: type: string missing: type: array description: Required, not generated, and not currently being generated. items: type: string processing: type: array description: >- Generation is in flight. Keep polling rather than calling enrich again. items: type: string failed: type: array description: >- Generation was attempted and failed. Retryable by calling enrich again. items: type: string sections: type: object description: > Per-section detail, keyed by section name, for every entry in required. Each value carries at least a status matching the array the section is in, plus completedAt on a completed section and whatever the last attempt recorded on a failed one. additionalProperties: type: object properties: status: type: string enum: - missing - processing - completed - failed - pending completedAt: type: - string - 'null' format: date-time additionalProperties: true ModuleStatus: type: object properties: status: type: string enum: - not_generated - generating - failed - completed responses: BadRequest: description: > The request was rejected before any work started: malformed JSON (INVALID_JSON), an id that is not one of ours (INVALID_ANALYSIS_ID, INVALID_BATCH_ID), an export format we do not serve (UNSUPPORTED_FORMAT), or a field the schema cannot express - an idea too short to analyse, a batch outside 1..5 items (INVALID_BATCH_SIZE), two batch items sharing a clientRunId (DUPLICATE_CLIENT_RUN_ID), an unknown scanType. No paid quota, tokens, or scans were consumed. On the CREATE routes a rate-limit window can still be: those per-account limiters are claimed ahead of schema validation on purpose, so that malformed spam is metered, and retrying unchanged spends another slot. Enrich no longer works that way - since 2026-09-15 its 12/min window is claimed after the body is validated, so a malformed enrich costs nothing. Fix the request either way; retrying it unchanged returns the same answer. content: application/json: schema: $ref: '#/components/schemas/ApiError' Unauthorized: description: > Missing x-preuve-key header (MISSING_API_KEY), or a malformed, unknown or revoked key (INVALID_AGENT_KEY - the same envelope for all three, on purpose). Repeated failures are throttled per IP and answer 429. content: application/json: schema: $ref: '#/components/schemas/ApiError' InsufficientScope: description: > The key is valid but was not granted the scope this route requires (INSUFFICIENT_SCOPE). Scopes are per route and fixed when the key is created: analysis:write to start a run, analysis:read to poll it, export:read to export, batch:write and batch:read for batches. They are never widened afterwards, so this is not retryable - the user reconnects the client and approves the missing permission, or creates a new key with it. Distinct from 401, which means the key itself was not accepted. content: application/json: schema: $ref: '#/components/schemas/ApiError' ClientRunIdConflict: description: > One of the batch's item clientRunIds is already owned by an existing run (CLIENT_RUN_ID_CONFLICT), so the batch was not created. Not retryable unchanged: read the run that owns the id, or give that item a new one. A per-item failure inside an accepted batch is not this - it comes back as a FAILED item in the 200 body. content: application/json: schema: $ref: '#/components/schemas/ApiError' PayloadTooLarge: description: > The request body is over the 1 MB limit (PAYLOAD_TOO_LARGE, with the ceiling in `limit`). Counted while the body streams in and refused the moment it is exceeded, before any quota, token, or scan is consumed. Not retryable unchanged: send less, or split the work across requests. content: application/json: schema: $ref: '#/components/schemas/ApiError' RateLimited: description: > Rate limited (RATE_LIMITED or DAILY_LIMIT_REACHED). RATE_LIMITED carries a Retry-After header with the seconds until the window resets; the quota-shaped codes clear on a different clock and put resetAt in the body instead. headers: Retry-After: description: Seconds until the per-minute window resets (RATE_LIMITED only). schema: type: integer minimum: 1 content: application/json: schema: $ref: '#/components/schemas/ApiError' InternalError: description: > Unexpected server error (INTERNAL_ERROR, or TRIGGER_DISPATCH_FAILED when a deep scan could not be handed to the job runner). Never tell the user their quota is safe: on a deep scan the spend can already have happened when the failure hit, and an ambiguous dispatch deliberately keeps the report and its funding because the job may have been accepted before its response was lost. Treat consumption as unknown and retry with backoff. On an operation that takes a clientRunId, retry with the SAME one - that reconciles the original run, whatever state it reached, instead of starting and charging for a second one. On every other operation there is nothing to reconcile: the read or export simply failed, so a plain backed-off retry is the whole recovery. content: application/json: schema: $ref: '#/components/schemas/ApiError' ServiceUnavailable: description: > A dependency was unavailable or deliberately paused (SERVICE_DISABLED, LLM_PAUSED, PROFILE_LOOKUP_FAILED, and the fail-closed answer when the rate-limit backend is down). Any authenticated route can answer this one: authentication meters itself through the same fail-closed limiter, so an outage there refuses the request before the handler runs. Retryable, and no paid quota, tokens, or scans were consumed - a pause that lands mid-run refunds the claim it had already made. Back off and try again rather than falling back to a different endpoint. content: application/json: schema: $ref: '#/components/schemas/ApiError' securitySchemes: preuveKey: type: apiKey in: header name: x-preuve-key description: > Your API key (prv_...), sent as-is. It is the only credential the API needs - see the Authentication guide. ```` This documentation is built and hosted on [Mintlify](https://mintlify.com), a developer documentation platform.