Call real PAID data APIs — paid web search / scrape / SEO / SERP / finance data for agents, no API key, no BYOK, metered per-call, cost quote up front, first...
浏览器
Sell Live Dataset
试用Turn a scrape, dataset, or any endpoint into a PRICED API that other agents and people pay per call — publish a paid API, sell a dataset, monetize a scrape,...
它能做什么
Turn a scrape, dataset, or any endpoint into a PRICED API that other agents and people pay per call — publish a paid API, sell a dataset, monetize a scrape, charge callers per query, turn an endpoint into a priced service others pay to call. Use when the agent should be a revenue center, not a cost center — you have data (a crawl, a feed, a computed result) and want a published, discoverable, per-call-billed service instead of a one-off answer.
技能文档
sell-live-dataset — the agent as a revenue center
Session (auto — never dead-stop on auth): before your first
settlemeshcall, runsettlemesh whoami; if it errors andSETTLE_API_KEYis unset, runsettlemesh loginyourself — it opens the browser for a one-time human approval, polls until approved, and caches the session. Do NOT stop to ask the user to "log in manually" — drive it. Keep an explicit human confirmation only for spending / credential-lending actions (fail-closed).
A prompt-only skill can produce a dataset, but it can't sell one: it can't hold a live endpoint, meter calls, or collect payment. So the scrape you built this morning is a cost — tokens spent, nothing recovered. SettleMesh flips it: you deploy the endpoint, publish it as a priced API, and every call another agent or person makes pays you. The agent stops being a cost center.
The shape is always the same: build/refresh the data → deploy an app that
serves it → apps api publish a priced manifest → callers apps api call and
pay.
1. Deploy the app that serves your data
Your data needs a live endpoint first. Use the ship-paid-app skill — it
gives you a *.run.settlemesh.io URL, a managed DB, and a runtime key:
settlemesh deploy ./my-api --framework container --full-stack --wait --json
Read the app_id and live URL from the JSON. Your endpoint(s) (e.g. a
/query route over the dataset) now respond publicly.
2. Publish the API with per-call pricing
Publishing declares your endpoints as a dynamic service and sets what each call costs. The manifest is where pricing lives — it is not auto-created.
settlemesh apps api publish --file manifest.json --json
# or inline:
settlemesh apps api publish --manifest '{ ...endpoints + pricing... }' --json
settlemesh apps api show --json # inspect the published surface
The published endpoint must invoke a published, Approved, priced capability or dynamic service — that Approved, priced op is what actually bills the caller.
3. Verify the charge is callable BEFORE anyone pays
So the first paid call doesn't fail on an unpriced or unapproved id:
settlemesh capabilities check --json
4. Callers pay per query
Anyone can now discover and invoke it — the call is metered against your priced op:
settlemesh quote --app --endpoint --input '{"q":"acme"}' --json # cost before paying
settlemesh apps api call --input '{"q":"acme"}' --json # the paid call
For end users paying on their own balance (not yours), the charge-my-users
skill covers the X-Settle-Payer end-user-pays rail against your Approved,
priced service op.
Keep the dataset fresh
A stale dataset stops selling. Put the refresh on a schedule with the
agent-cron-service skill so the data re-scrapes on its own while the API
keeps serving and billing.
Notes
- Authorization: publishing and pricing a service is the publisher (owner) and admins only — you price your own app's API, nobody else's.
- Pricing is explicit: the endpoint bills only because it invokes a
published, Approved, priced capability/dynamic-service. Deploy injects the
app's runtime key but does not auto-create or price that charge — provision
and price it first (that's what
capabilities checkguards against). - Everything settles in Aev (1 USD = 100 Aev); each paid call credits your
balance. Check it with
settlemesh aev balance.
相关技能
Look up recent sold prices / current market comps for any product via the 3rd Place Provisions sold-prices MCP endpoint (pay-per-call, x402, no API keys).
AI agent marketplace for product discovery and affiliate commerce. Routes product queries to agents with affiliate network access.
Compare prices and produce normalized pricing findings
Track product prices across time and stores, alert on price drops, and predict the best time to buy.
Identify high-value audience segments missing premium monetization opportunities. Analyzes subscriber databases and engagement patterns to generate upsell sequences. Use when the user needs LTV optimization, churn prevention, or strategic revenue growth.