| name | kg-ask |
| skill_type | skill |
| description | Answers a natural-language question over the Knowledge Graph by generating and executing a query (Cypher/SQL/SPARQL) for you. Use when you want KG data but would rather ask in plain English — "ask the graph…", "how many…", "list the… " without writing Cypher. |
| license | MIT |
| tags | ["graph-os","query","nl"] |
| tier | core |
| wraps | ["graph_ask","ask_data"] |
| metadata | {"author":"Genius","version":"0.1.0"} |
kg-ask
Condensed intent-surface note (Seam 8). Under the default intent surface (MCP_TOOL_MODE=intent), graph_ask, ask_data are held back from the default tool list (nothing removed — REST + _execute_tool still reach them exactly as documented below). Two ways to use this skill unchanged: (1) load_tools(tools=["graph_ask"]) once per session (as below), then proceed exactly as documented; or (2) call the ask intent verb with the same natural-language request — the resolver routes to graph_ask for you and returns the result plus a routing justification. Set MCP_TOOL_MODE=condensed/verbose/both to expose the granular tools eagerly instead.
graph_ask turns a plain-English question into a read-only query and runs it, returning the generated query (auditable) plus result rows. dialect is auto by default (the model picks) or force cypher/sql/sparql; set execute=false to preview the generated query without running it.
Invoke
- MCP:
load_tools(tools=["graph_ask"]), then graph_ask(question="How many Concept nodes exist?").
- REST twin:
POST /graph/ask with {"question": "...", "dialect": "auto", "execute": true, "limit": 50}.
Example
graph_ask(question="Which agents depend on the mcp-multiplexer?", execute=true)
