The Preuve AI startup validation API lets any AI agent validate startup ideas programmatically. The Agent API and its MCP server are live. The remote MCP server at https://mcp.preuve.ai/mcp works as a claude.ai custom connector with OAuth sign-in, or with Claude Code, Cursor, Codex, and any HTTP MCP client; a local stdio server is also available. The API runs the Preuve analysis and returns a viability score, market research, competitor analysis, a structured ideas-json export, and a shareable report URL. Deep reports can generate optional modules: Proof of Demand, Founder Fit, Playbook, Trends, and Community Signals. A ready-made Agent Skill at docs.preuve.ai/agent-skill teaches AI agents the full Preuve workflow, including quota safety and retry semantics. API keys never grant access to the Preuve dashboard or account settings.
MCP · Agent API
Validate the idea before your agent builds it.
Your agent ships in minutes, but only 17.5% of ideas earn a go. Preuve runs the full, source-linked viability check first - from inside your agent.
Claude Code
Cursor
Codex- 7,000+
- ideas analyzed
- 17.5%
- earn a go-verdict
- 60+
- live sources per scan
- ~10 min
- to a scored verdict
Connect your agent
One paste connects your agent to all eight Preuve tools - pair it with the Agent Skill. On claude.ai, the fastest path is one click from the Claude Connectors Directory.
Get your API key
Start free: create your Preuve account, connect from Claude, and run your first starter scans at no cost. The full Founder Report ($29 one-time) unlocks 18-section deep analysis.
Create your accountAlready have one? Sign in and come back.
# One click - OAuth sign-in, no key, works on a free account claude.ai/directory/preuve-ai
1. Open the directory listing → Connect
2. Sign in with Preuve → approve
# 8 tools appear: start_analysis, get_analysis, enrich_analysis, ...
# Or manually: Settings → Connectors → Add custom connector → https://mcp.preuve.ai/mcp
In your Claude chat
What to ask once it's connected
Paste any of these into claude.ai, Claude Code, or Cursor. Claude picks the right Preuve tool on its own.
Startup validation API
Available now. This API-key-authenticated Agent API creates analysis runs (no-cost Starter scans or full paid deep scans), polls them, generates the export sections and optional deep modules, and returns the same analysis a human user gets - as structured JSON.
Endpoints
Connecting from Claude mints a key on any plan, free included. Creating one by hand in the account panel is self-serve for paid accounts. Every key carries explicit scopes, and analysis usage is charged to the owning account quota - Starter scans at no cost, deep scans from subscription, lifetime, or token balance.
/api/agent/analysesanalysis:writeStart one analysis (starter or deep)/api/agent/analyses/:idanalysis:readPoll one analysis/api/agent/analyses/:id/enrichanalysis:writeGenerate export sections + deep modules/api/agent/analyses/:id/exportexport:readExport one analysis as JSON/api/agent/analysis-batchesbatch:writeStart up to 10 analyses/api/agent/analysis-batches/:idbatch:readPoll a batch/api/agent/analysis-batches/:id/exportexport:readExport a batch as JSON/api/agent/tools/idea-generatoranalysis:writeGenerate startup ideas (backs generate_ideas)Auth
One header: x-preuve-key, holding the whole key. No signing, no timestamps, no nonces.
curl https://preuve.ai/api/agent/analyses \
-H "x-preuve-key: $PREUVE_API_KEY" \
-H "content-type: application/json" \
--data '{"clientRunId":"run-online-001","scanType":"starter","idea":"A workflow automation service for solo founders that validates market demand before they build.","publish":true}'Node client
Keep PREUVE_API_KEY server-side. The key carries no origin restriction, so it never belongs in browser code.
async function preuve(path, { method = 'GET', body } = {}) {
return fetch(`https://preuve.ai${path}`, {
method,
headers: {
'content-type': 'application/json',
'x-preuve-key': process.env.PREUVE_API_KEY,
},
body: body ? JSON.stringify(body) : undefined,
});
}Scopes
Scopes are exact. Write access for analyses does not grant export access.
analysis:writeCreate analysis runs and enrich them for the key owneranalysis:readPoll single-run status, module states, and URLsbatch:writeCreate bounded batches up to 10 ideasbatch:readPoll batch status and child run URLsexport:readExport completed analyses and batches as JSONResponses
Polling returns status, enrichment progress, per-module states, and URLs. The export endpoint returns the structured report itself (ideas-json, schemaVersion 2) - full deep sections included on paid runs.
{
"id": "run_123",
"reportId": "report_123",
"clientRunId": "run-online-001",
"scanType": "deep",
"analysisTier": "advanced",
"status": "COMPLETED",
"readyForExport": true,
"enrichment": { "mode": "core", "status": "completed" },
"modules": {
"proofOfDemand": { "status": "completed" },
"founderFit": { "status": "not_generated" },
"playbook": { "status": "generating" },
"trends": { "status": "not_generated" },
"community": { "status": "not_generated" }
},
"reportUrl": "https://preuve.ai/app/results?id=report_123",
"shareUrl": "https://preuve.ai/share/abc123"
}Deep modules
Paid deep reports can generate five optional modules through the enrich endpoint. One successful generation per module per report, included in the deep scan price - a failed generation never consumes the cap.
proofOfDemandVerified real-demand prospects with verbatim quotes. Beta: results can be uneven, every quote links to its source.asyncfounderFitFounder-vs-plan underwriting from a submitted founder profile: fit score, top risks, pivot fit.inlineplaybookA concrete launch playbook generated for the validated idea.asynctrendsGoogle Trends data for the idea, generated on demand - the analysis never fetches it on its own.asynccommunityCommunity signals: real discussions from forums, Reddit, review sites, and X about the market, generated on demand - the analysis never fetches it on its own.asyncStatus lifecycle
Batches aggregate child runs into PROCESSING, COMPLETED, PARTIAL, or FAILED.
PENDINGThe run record exists and is waiting for dispatchPROCESSINGThe Preuve analysis pipeline is runningCOMPLETEDThe report is done - enrich if needed, then exportFAILEDThe run reached a terminal error with a stable codereportUrl
/app/results?id=<reportId>Private owner dashboard
shareUrl
/share/<shareId>Public shared dashboard
ideas-json
schemaVersion: 2Structured export for single runs and batches
What a key can do - and what it can't
An API key is a scoped analysis credential, not a login. It shares your account quota, and nothing else about your account.
Included
- Run analyses. No-cost Starter scans (Reality Checks) and full paid deep scans, drawn from your account quota.
- Poll and enrich. Run status, per-module states, idempotent enrichment of export sections.
- Structured exports. Stable ideas-json (schemaVersion 2): scores, research, and the full deep report body on paid runs.
- Deep modules. Proof of Demand, Founder Fit, Playbook, Trends, and Community Signals on your deep reports.
- Share URLs. Optional public share links when you pass publish: true.
Not included
- No dashboard access. A key cannot log into preuve.ai, read your account, or open the dashboard. It is not a session.
- No account or billing control. Keys cannot change plans, buy tokens, manage payment methods, or create/revoke other keys.
- API-created runs only. A key only sees analyses it created. Reports you ran in the web app stay invisible to it.
- No profile writes. A founder profile submitted for Founder Fit is scoped to that report - it never rewrites your account preferences.
Startup validation API FAQ
What is the Preuve startup validation API?
It is an HTTP API that runs a Preuve analysis on a startup idea and returns a viability score, market research, competitor analysis, a shareable report URL, and a structured JSON export. You authenticate each request with a single API key header, so any AI agent, CLI, or backend script can validate ideas programmatically.
Is there a Preuve MCP server?
Yes - the MCP server is live with two transports and the same eight tools (start_analysis, get_analysis, enrich_analysis, export_analysis, create_batch, get_batch, export_batch, generate_ideas), and listed in the Claude Connectors Directory (claude.ai/directory/preuve-ai). The remote server at https://mcp.preuve.ai/mcp works as a claude.ai connector (OAuth sign-in and a consent screen, no key handling) or with any HTTP MCP client sending an Authorization header; the original local stdio server remains fully supported. A ready-made Agent Skill (docs.preuve.ai/agent-skill) teaches your agent the full workflow: scan-type quota safety, polling, enrichment gates, and safe retries. Failed runs refund what they claimed, and error codes are stable so agents can retry safely.
How do I get a startup validation API key?
Two ways. Connecting the Preuve connector from Claude signs you in and mints the key for you during consent, on any plan including free. Creating a key by hand in the account panel is self-serve for paid accounts. Start free: connect and run your first starter scans at no cost. The full Founder Report ($29 one-time) unlocks 18-section deep analysis. Each key is scoped (analysis:write, analysis:read, batch:write, batch:read, export:read), sent in one x-preuve-key header, and tied to one Preuve account.
What is the difference between a starter and a deep scan?
scanType is explicit on every run. "starter" runs the Reality Check at no cost from your Starter scan allowance. "deep" runs the full paid analysis (15+ sections, pivots, citations) and consumes your account quota: subscription scans first, then lifetime monthly scans, then tokens. An agent can never spend paid quota by accident, and a refused deep scan fails loudly instead of silently downgrading.
What does the API return?
Polling returns run status, enrichment progress, per-module states, and URLs. The export endpoint returns structured ideas-json (schemaVersion 2): scores, verdict, market, competitors, SWOT, risks, quick take, and citations on every tier, plus the full deep report body (sections, pivots, and generated module payloads) on paid deep runs owned by your key.
What are deep modules?
Optional add-on generations on paid deep reports, requested through the enrich endpoint: proofOfDemand (verified real-demand prospects), founderFit (founder-vs-plan underwriting from your founder profile), playbook (launch playbook), trends (on-demand Google Trends), and community (community signals from forums, Reddit, review sites, and X). Each module is capped at one successful generation per report and is included in the deep report price - no extra fee.
Does an API key give access to my Preuve dashboard or account?
No. A key is not a login: it cannot open the dashboard, read your account, change billing, or manage other keys. It only sees analyses it created through the API - reports you ran in the web app stay invisible to it. If a key leaks, revoke it; your account itself is untouched.
How much does the validation API cost?
API usage draws from the same analysis quota as your Preuve plan - Starter scans from your Starter allowance at no cost, deep scans from your subscription, lifetime, or token balance. There is no separate API fee, and failed runs refund whatever they claimed.
Should I validate before using an autonomous agent like Polsia?
Yes. Autonomous build agents such as Polsia execute fast but skip the demand-validation step, which is where most teams burn their runway. Run the idea through the Preuve validation API first, then hand a validated brief to your build agent. See the linked Polsia review for a deeper breakdown of that gap.
New to Preuve? See how the analysis works on the idea validation page, browse real example reports, or check pricing and scan limits. Weighing an autonomous build agent? Read our honest Polsia review on why validation has to come first. The MCP server is listed in the Claude Connectors Directory, on the official MCP registry and Smithery, cursor.directory, and mcpservers.org.