AgensGraph Cypher MCP server (schema introspection + read/write Cypher tools).
Listed on its own since 2026-09-26. Until then it was part of mcp-agensgraph-common, where its earlier checks remain.
Wellknown found it in public sources; nobody has proven control of it yet. Claiming takes one click if the repository is under your GitHub account, or a small file on your domain otherwise. Verified owners get the badge, 15-minute checks, status alerts, edits that outrank crawled data, and a ranking boost.
Agents can do it too: POST https://wellknown.network/api/v1/claims with {"agent":"mcp-agensgraph-cypher","method":"well_known_file"} — machine-readable steps at claim.json, guide at /docs/claim.
Everything here was measured by our prober or read from a registry. Nothing is self-reported.
Attributed to the source that supplied each field. Treated as claims, not facts.
# 🔍⁉️ Agensgraph MCP Server ## 🌟 Overview A Model Context Protocol (MCP) server implementation that provides database interaction and allows graph exploration capabilities through Agensgraph. This server enables running Cypher graph queries, analyzing complex domain data, and automatically generating business insights that can be enhanced with Claude's analysis. ## 🧩 Components ### 🛠️ Tools The server offers these core tools: #### 📊 Query Tools - `read_agensgraph_cypher` - Execute Cypher read queries to read data from the database. Runs in a read-only transaction (the database rejects writes even if the keyword check is bypassed), and results are **paginated** so an unbounded query can't flood the context. - Input: - `query` (string): The Cypher query to execute - `params` (dictionary, optional): Parameters to pass to the Cypher query - `limit` (int, optional): max rows for this page (default 100; clamped to a server max, default 1000) - `offset` (int, optional): rows to skip — pass the response's `next_offset` to fetch the next page. Skipping produces and discards the rows before the page, so the cost grows with the offset: measured over 50,000 nodes, offset 0 took 0.062 s, 10,000 took 1.19 s and 40,000 took 6.21 s. - `walk` (bool, optional): read the whole result in one pass from a server-side cursor instead of a page of it, and report the exact `total_rows`. Paging through a whole result costs the sum of the offsets: measured over 200,000 rows, pages of 1,000 took 7.90 s and one walk took 0.70 s. - Returns: a JSON object `{ "rows": [...], "row_count", "offset", "limit", "has_more", "next_offset" }`. When `has_more` is true, call again with `offset = next_offset` for the next page. (Page sizes are configurable via `AGENSGRAPH_PAGE_SIZE` / `AGENSGRAPH_MAX_PAGE_SIZE`.) A response is bounded by a token budget — 10,000 tokens by default, set with…
Mapped onto the structured taxonomy from declared text and observed tool names. Confidence shown for derived entries.
Every source is kept verbatim. Field changes are logged as events.