conviva-mcp.yml
Conviva operates TWO hosted, remote-only Model Context Protocol servers. Neither can be run locally. Both are OAuth 2.1 protected resources implementing RFC 8414 + RFC 9728 discovery with dynamic client registration and PKCE (S256); mcp.conviva.com additionally accepts HTTP Basic with a Pulse API key pair for autonomous (non-interactive) agents. A live tools/list on both servers returns JSON-RPC error -32001 "Unauthorized" (HTTP 401), so tool input schemas require an authenticated introspection. The tool names below are NOT guessed — they are taken verbatim from Conviva's own published Agent Skills in github.com/Conviva/mcp-marketplace, mirrored in this repo under skills/.
One-click install for Cursor, VS Code, Claude, and 20+ other MCP clients, powered by API Commons MCP Install — visit install.apicommons.org for more information.
Documentation
Documentation link · transport
Tools
identity-c3-account-list— List the c3 accounts the caller is authorized for. Account-scoped tools require an explicit c3AccountName; there is no safe default.context-center-nexa-semantic-search— Vector/ANN semantic search over the Context Center knowledge layer; accepts a free-text query, optional exact-substring terms, and categories scoping; returns hits with a similarity score.context-center-nexa-categories-list— List the valid knowledge categories (asset types) usable for scoping semantic search.context-center-nexa-knowledge-get— Fetch a knowledge entry by its k_* id — the reverse lookup from an asset's knowledge_id.context-center-asset-get— Hydrate the authoritative structured definition of one Context Center asset by assetType + assetId (pattern, metric, critical-event, dimension, segment, template).context-center-asset-list— Page through all assets of an assetType, returning { items, pagination }. The search argument filters one already-fetched page, not the whole result set.metric-query-run— Run a predefined metric by metricId over a startDate/endDate window, optionally with a patternId and groupByDimensionIds. Accepts ids only — never raw SQL or a raw query payload.nexa-analyze— Async open-ended analysis fallback for questions no predefined metric answers. Requires a fallbackReason (no_matching_metric, unsupported_breakdown, open_ended_analysis, multi_metric_or_comparison, metric_query_failed, other).nexa-analyze-result— Retrieve the result of an async nexa-analyze job. Backed by a Redis job store.insights-behavior-segment-list— List Insights behavior segments (formerly "patterns") by name.insights-behavior-segment-get— Retrieve the stored details of one Insights behavior segment — its pre-computed insights, evidence and conversion barriers. A static lookup, not a live analytical query.
About MCP
The Model Context Protocol (MCP) is an open protocol Anthropic introduced for connecting LLM-based agents to external tools and data sources. Providers publish MCP servers that expose their API surface as structured, discoverable tools — an MCP-compatible client (Claude Desktop, Cursor, Cline, Continue, etc.) can connect to the server and call its tools without any per-provider integration code.
Browse every MCP server on the APIs.io network or compare with the broader Agent Skill surfaces of the same providers.
Work with this as data
Every MCP server here is available over the APIs.io API and to AI agents over MCP.