Xpoz MCP
A social-data layer for agents
Xpoz is a social-data service for developers and AI-assisted research workflows. Its official overview lists Twitter/X, Instagram, TikTok, and Reddit, with access through a remote MCP server, TypeScript and Python SDKs, and a CLI. The documented tasks include searching posts, examining profiles, filtering engagement, and monitoring selected accounts or terms.
The publisher explicitly describes Xpoz as a data layer rather than an all-in-one social-listening dashboard. A team can use its data in its own alerts, monitoring, or analysis pipeline; it should not assume that a prebuilt business dashboard is included.
The homepage's no-platform-API-key claim concerns the source networks. Xpoz itself documents account authentication and access keys, so it is not an unauthenticated unlimited data feed.
Search and continuous monitoring
The service advertises filters for date, author, language, and engagement, along with field selection, caching, and pagination. Tracked keywords and users are described as a way to collect matching new posts continuously and make them available through the tools.
These features can support a defined research question or brand-monitoring process. They do not establish that the captured corpus covers every account, deleted post, historical period, or platform event. Keep the query, dates, filters, and returned source references with an analysis so its scope remains inspectable.
Sentiment and influence features are advertised, but labels and scores remain interpretations. Review the original posts before making consequential claims about people or organizations. Use collected data only within your authorization and the applicable source and service terms.
Query credits and account limits
The captured pricing page states that credits are charged per query according to the platform, not per returned row. A large result set therefore has a different pricing basis from a service billed by record, though other plan and query limits still apply.
The free tier provides a one-time allocation of 500 credits that does not refresh monthly. Paid plans advertise monthly credits and larger tracked-keyword or user allowances, with overage charges. The visible paid prices are annual-billing equivalents; check the selected billing period rather than treating them as a month-to-month payment.
For evaluation, use a small public-data query with a clear purpose, inspect the returned fields and source references, and check credit consumption. This overview has not run the API or verified claims about corpus size, completeness, processing speed, or real-time coverage.