Periskop – product discovery (MCP/API) for AI agents

duarteapereira1 pts0 comments

Periskop — Product discovery for AI agents<br>View docsRequest MCP/API access

For developers · MCP / API<br>Product discovery for AI agents.<br>Cheap, reliable, hallucination-free product discovery for the agents you're building. One MCP/API call turns shopping intent into real, in-stock products, with live prices and working merchant links.<br>Request MCP/API accessView docsTalk to us<br>Soon also on

browsebundlerecommendation<br>run_shopping_discovery<br>200 OK<br>prompt ▸ "▋"

Example output · hover to pause, or pick an answer_type

MCP / API live<br>Integrate today<br>Paste-ready agent reply<br>suggested_agent_reply<br>Honest by default<br>match_quality + caveats<br>No checkout, ever<br>Merchant links only

MCPREST<br>MCP for agents<br>// MCP, Claude Code, Cursor, custom agent runtimes<br>"mcpServers": {<br>"periskop": {<br>"url": "https://mcp.periskop.ai",<br>"headers": { "Authorization": "Bearer dp_..." }

REST for backends<br>// REST, from any app backend<br>POST /shopping/discover<br>Authorization: Bearer dp_...<br>{ "prompt": "...", "mode": "bundle" }

In the developer portal<br>API keysUsageBillingDocsOpenAPI<br>Create a project, mint a key, and call run_shopping_discovery. Every result carries a result_id, re-fetch it with get_discovery_result.<br>Works with<br>Drop Periskop into your existing automation stack, native steps for Dify, n8n, and Pipedream.<br>Difyn8nPipedream

Request MCP/API accessView docs

Without Periskopevery request<br>prompt ▸ "▋"<br>01web_search×1· no strong matchcost<br>02open page· stale / out of stockcost<br>03web_search· broaden terms · still thincost<br>04open page×1· 404 · dead linkcost<br>05llm reads it all· reads everythingcost

Latency

unpredictable<br>Reliability

low ↓

→ still no reliable product

N steps→1 call

With Periskop · one callrenderedjson<br>prompt ▸ "cabin bag that fits Ryanair's size limit, under €60"<br>run_shopping_discovery→ 200 OK<br>answer_type recommendation · summary 2 in-stock picks within 40×20×25cm<br>Cabin Max Metz (40×20×25)amazon.frEUR 39.99<br>Aerolite Ryanair Cabin Bagdecathlon.frEUR 27.99<br>suggested_agent_reply<br>Two cabin bags under €60 that fit Ryanair's 40×20×25 cm limit: the Cabin Max Metz (€39.99) and the Aerolite Ryanair bag (€27.99), both in stock. Want the cheaper one, or a hard-shell option?

next_actionsCheaperHard-shell only

Latency

one call<br>Reliability

high ↑

ranked, in-stock, linked, ready to show

Swipe to compare →<br>Same retrieval, ranking, and validation, run once, ahead of time, not re-derived on every request.

1 · Store attempted<br>Accessible<br>Product pages found<br>Product cards detected

2 · Validated by Periskop<br>Product extraction<br>Price readability<br>Link validity<br>Image presence<br>Stability / latency

3 · Allowed use<br>Trusted , can shape best picks<br>Limited , alternatives or restricted contexts<br>Rejected , cannot influence recommendations

Validated, discovery-ready merchant supply is what makes agent recommendations safer.

The hard part nobody solved, Periskop<br>Intent understanding<br>Retrieval<br>Ranking<br>Alternatives<br>Bundles<br>Caveats<br>Match quality<br>Merchant links

The easy part, already solved<br>Stripe · merchants · your storefront<br>Checkout<br>Payments<br>Orders<br>Fulfillment<br>Returns

One call · imagination is the limit<br>AI shopping assistantsBrowser & computer-use agentsPrice & stock monitoringArbitrage & resale sourcingProcurement & opsCommerce copilotsAI-native shopping appsAccessibility shopping+…and whatever you ship next

AI shopping assistants<br>via run_shopping_discovery<br>Real, in-stock picks dropped straight into chat, ranked, priced, and linked.<br>wireless earbuds under €50, good for running<br>Two solid picks under €50, both sweat-resistant:JBL Wave Buds<br>amazon.fr

€34.99<br>open

Anker Soundcore P20i<br>fnac

€29.99<br>open

ranked from 40+ in-stock · live prices

Usage-based · €7 / 1,000 shopping requests · free credits on your first keySee pricing →

Build with Periskop<br>Product discovery for the agents you are building.<br>For AI shopping assistants, browser and computer-use agents, procurement workflows, commerce copilots, recommendation systems, and AI-native shopping apps.<br>Request MCP/API accessView docsTalk to us

What is Periskop?+Periskop is product discovery for AI agents. One MCP/API call turns a shopping request into real, in-stock products, with live prices, working merchant links, alternatives, inferred bundles, caveats, a paste-ready reply, and a result_id, so your agent can recommend instead of crawling the web on every request.<br>Who is this for?+Developers building AI shopping assistants, browser and computer-use agents, procurement and ops workflows, commerce copilots, recommendation features, accessibility shopping, and AI-native shopping apps.<br>How does it work?+You send a prompt to run_shopping_discovery over MCP or REST. Periskop retrieves and ranks across quality-gated merchants and returns a structured response, items or an inferred bundle, plus a paste-ready suggested_agent_reply and next_actions, your agent renders directly. No per-site browsing, no runtime scraping.<br>Do merchants need to integrate?+No. Periskop validates store structure,...

periskop shopping product agents stock discovery

Related Articles