CWI Discovery Engine

Get Catalog Graph

get_graph

Read-onlyIdempotent

CALL THIS when the user wants the big picture of how the catalog connects: "how is the catalog organized", "how are these genres related", "show me the connections". Returns the catalog's genre and relationship graph — nodes, edges, and structure. Use get_genre instead when the user asks what one specific genre is.

Tool call template

This tool declares no required parameters. Add optional parameters from the input schema as needed.

{
  "id": 1,
  "jsonrpc": "2.0",
  "method": "tools/call",
  "params": {
    "arguments": {},
    "name": "get_graph"
  }
}

Input Schema

Name	Required	Description	Default
No arguments

Output Schema

Name	Required	Description
`source`	Yes
`_session`	No	Chain session: pass _session.session_token back in the next call to continue the chain.
`try_next`	No	Suggested follow-up calls, each a pre-compiled {tool, arguments, purpose} payload.
`full_graph`	Yes
`generated_at`	No

Schema Changelog

Changes observed during successful MCP inspections.

Changed1 schema field changedOct 15, 2026

changedOutput schema / (root)

Previous value: -nullNew value: +{
+  "additionalProperties": true,
+  "description": "get_graph response: catalog knowledge-graph summary (node/edge counts, samples), not the full graph.",
+  "properties": {
+    "_session": {
+      "additionalProperties": false,
+      "description": "Chain session: pass _session.session_token back in the next call to continue the chain.",
+      "properties": {
+        "session_token": {
+          "description": "Hex session token; echo it as params._session.session_token to continue.",
+          "type": "string"
+        },
+        "step": {
+          "description": "Hop number in this chain (1 = chain start).",
+          "type": "integer"
+        }
+      },
+      "required": [
+        "session_token",
+        "step"
+      ],
+      "type": "object"
+    },
+    "full_graph": {
+      "type": "string"
+    },
+    "generated_at": {
+      "type": "string"
+    },
+    "source": {
+      "type": "string"
+    },
+    "try_next": {
+      "description": "Suggested follow-up calls, each a pre-compiled {tool, arguments, purpose} payload.",
+      "items": {
+        "additionalProperties": false,
+        "description": "Pre-compiled next call: copy `arguments` straight into the next tools/call.",
+        "properties": {
+          "arguments": {
+            "additionalProperties": true,
+            "description": "Ready-to-use arguments, pre-populated from the current response.",
+            "type": "object"
+          },
+          "purpose": {
+            "description": "Why this call is the natural next step.",
+            "type": "string"
+          },
+          "tool": {
+            "description": "Tool to call next.",
+            "type": "string"
+          }
+        },
+        "required": [
+          "tool",
+          "arguments"
+        ],
+        "type": "object"
+      },
+      "type": "array"
+    }
+  },
+  "required": [
+    "source",
+    "full_graph"
+  ],
+  "type": "object"
+}

First observedOct 5, 2026

TDQS

A4.4/5.0

Behavior3/5

Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?

Annotations already declare readOnlyHint, idempotentHint, and destructiveHint, so the safety profile is fully covered. The description only restates that a graph structure comes back, which overlaps the output schema, and says nothing about graph size, pagination, or cost constraints for a potentially large traversal.

Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.

Conciseness4/5

Is the description appropriately sized, front-loaded, and free of redundancy?

Front-loads the call condition before the return value and the alternative, which is the right ordering. Slightly padded by listing three example phrasings and a redundant 'nodes, edges, and structure' clause that repeats 'graph', but nothing is truly wasted.

Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.

Completeness5/5

Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?

For a parameterless read-only tool with a full output schema, the description covers purpose, trigger conditions, and the sibling alternative. Return-value detail is rightly delegated to the output schema, leaving no meaningful gap.

Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.

Parameters4/5

Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?

The tool takes zero parameters, so there is nothing for the description to disambiguate and the baseline of 4 applies. No parameter semantics are needed or missing.

Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.

Purpose5/5

Does the description clearly state what the tool does and how it differs from similar tools?

States a specific verb and resource ('Returns the catalog's genre and relationship graph') and spells out the payload as nodes, edges, and structure. It also names the sibling get_genre as the wrong tool for single-genre questions, so an agent can separate the two without inspecting schemas.

Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

Usage Guidelines5/5

Does the description explain when to use this tool, when not to, or what alternatives exist?

Explicitly gives triggering user intents ('how is the catalog organized', 'how are these genres related', 'show me the connections') and an exclusion rule routing single-genre lookups to get_genre. This is a textbook when-to-use plus when-not-to-use statement.

Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.

All CWI Discovery Engine tools Inspect

Try in Browser

One endpoint for every MCP toolRayrun runs every MCP server behind one URL and holds the credentials. Each call is decided per client, tool, and argument, then recorded.ray.runAd

Glama MCP Gateway

Add one secure layer between your agents and this server.