MCP Canary

Inspect MCP endpoint

inspect_endpoint

Read-onlyIdempotent

Connects to a remote MCP endpoint (Streamable HTTP, then legacy SSE), lists its tools, and classifies each tool's side effects from its MCP hints. Never calls a tool on the target. agent_readiness lists what is wrong with the tool definitions and protocol compliance of the target, each with a fix, and a grade A-F. A target problem is status=failed with a stable failure code (e.g. TARGET_UNAVAILABLE, TARGET_AUTH_FAILED, MCP_PROTOCOL_ERROR). Private and loopback addresses are rejected. Set website (not url) to inspect a web page whose JavaScript registers WebMCP tools: the page is opened in a browser; the result has transport=webmcp.

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": "inspect_endpoint"
  }
}

Input Schema

Name	Required	Description
`url`	No	Absolute http or https URL of the MCP endpoint, e.g. https://mcp.deepwiki.com/mcp. Set url or website.
`headers`	No	Optional HTTP headers for the target, e.g. Authorization. Sent to the target only, never stored. Not with website.
`website`	No	Absolute http or https URL of a web page with WebMCP tools, e.g. a Shopify storefront. Set url or website.

Output Schema

Name	Required	Description	Default
`tools`	Yes
`server`	Yes
`status`	Yes
`failure`	Yes
`endpoint`	Yes
`protocol`	Yes
`transport`	Yes
`duration_ms`	Yes
`instructions`	No
`agent_readiness`	No

Schema Changelog

Changes observed during successful MCP inspections.

Changed8 schema fields changedNov 3, 2026

changedInput schema / properties / headers / description

Previous value: -"Optional HTTP headers for the target, e.g. Authorization. Sent to the target only, never stored."New value: +"Optional HTTP headers for the target, e.g. Authorization. Sent to the target only, never stored. Not with website."

addedInput schema / properties / url / default
```
Added value: +null
```

changedInput schema / properties / url / description

Previous value: -"Absolute http or https URL of the MCP endpoint, e.g. https://mcp.deepwiki.com/mcp"New value: +"Absolute http or https URL of the MCP endpoint, e.g. https://mcp.deepwiki.com/mcp. Set url or website."

changedInput schema / properties / url / type

Previous value: -"string"New value: +[
+  "string",
+  "null"
+]

addedInput schema / properties / website

Added value: +{
+  "default": null,
+  "description": "Absolute http or https URL of a web page with WebMCP tools, e.g. a Shopify storefront. Set url or website.",
+  "type": [
+    "string",
+    "null"
+  ]
+}

removedInput schema / required
```
Removed value: -[
-  "url"
-]
```
addedOutput schema / properties / tools / items / properties / hints / properties / consequential_hint
```
Added value: +{
+  "default": null,
+  "type": [
+    "boolean",
+    "null"
+  ]
+}
```
addedOutput schema / properties / tools / items / properties / hints / properties / untrusted_content_hint
```
Added value: +{
+  "default": null,
+  "type": [
+    "boolean",
+    "null"
+  ]
+}
```

Changed3 schema fields changedNov 1, 2026

addedOutput schema / properties / agent_readiness

Added value: +{
+  "default": null,
+  "properties": {
+    "errors": {
+      "type": "integer"
+    },
+    "findings": {
+      "items": {
+        "properties": {
+          "category": {
+            "type": "string"
+          },
+          "example": {
+            "default": null,
+            "type": [
+              "string",
+              "null"
+            ]
+          },
+          "fix": {
+            "type": "string"
+          },
+          "items": {
+            "default": null,
+            "items": {
+              "type": "string"
+            },
+            "type": [
+              "array",
+              "null"
+            ]
+          },
+          "message": {
+            "type": "string"
+          },
+          "rule": {
+            "type": "string"
+          },
+          "severity": {
+            "type": "string"
+          },
+          "tool": {
+            "default": null,
+            "type": [
+              "string",
+              "null"
+            ]
+          },
+          "why": {
+            "type": "string"
+          }
+        },
+        "required": [
+          "rule",
+          "category",
+          "severity",
+          "message",
+          "why",
+          "fix"
+        ],
+        "type": "object"
+      },
+      "type": "array"
+    },
+    "grade": {
+      "type": [
+        "string",
+        "null"
+      ]
+    },
+    "infos": {
+      "type": "integer"
+    },
+    "latest_protocol": {
+      "type": "string"
+    },
+    "score": {
+      "type": [
+        "integer",
+        "null"
+      ]
+    },
+    "warnings": {
+      "type": "integer"
+    }
+  },
+  "required": [
+    "grade",
+    "score",
+    "errors",
+    "warnings",
+    "infos",
+    "latest_protocol",
+    "findings"
+  ],
+  "type": [
+    "object",
+    "null"
+  ]
+}

addedOutput schema / properties / instructions

Added value: +{
+  "default": null,
+  "type": [
+    "string",
+    "null"
+  ]
+}

addedOutput schema / properties / tools / items / properties / title

Added value: +{
+  "default": null,
+  "type": [
+    "string",
+    "null"
+  ]
+}

First observedOct 24, 2026

TDQS

A4.3/5.0

Behavior5/5

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

Annotations already declare a safe, idempotent read, and the description goes well beyond them: it promises the target's tools are never invoked, discloses the transport negotiation order (Streamable HTTP then legacy SSE), names stable failure codes such as TARGET_UNAVAILABLE and TARGET_AUTH_FAILED, and explains that website mode opens a browser and returns transport=webmcp. These are exactly the behavior traits an agent cannot infer from structured fields.

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-loaded with the core action, then escalated to output shape and failure modes, then the special website path. Dense but nearly every clause carries information; the enumerated failure codes are the only slightly padded element.

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

Completeness4/5

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

Covers what an agent needs for a network-touching, open-world inspection tool: safety (no target tool invocation), address restrictions, transport fallback, result shape (agent_readiness, grade, status), and the alternate webmcp mode. Some operational details such as timeouts or authentication expectations for the target are left implicit.

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

Parameters3/5

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

Schema description coverage is 100% and each parameter already documents itself, including 'Set url or website' and 'Not with website', so the schema carries the parameter burden. The description reinforces the url-vs-website exclusivity in prose but adds little genuinely new semantics for headers or URL format.

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 sequence of verbs and resources: connects to a remote MCP endpoint, lists its tools, classifies side effects from MCP hints, and grades agent readiness. This is far more specific than the name alone and clearly distinguishes it from the only sibling, get_pricing.

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

Usage Guidelines4/5

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

Gives a concrete routing rule between the two entry paths: use website (not url) for a page whose JavaScript registers WebMCP tools, and implies url for a plain MCP endpoint. It also states exclusion criteria (private and loopback addresses are rejected). It does not mention when to prefer this over any other inspection approach, but the sibling set is tiny.

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

All MCP Canary tools Inspect

Try in Browser

Claude Code, inside the Godot 4 editorFlockbay puts an AI agent in the real Godot 4 editor: it writes GDScript, runs your game and fixes what breaks. Free, or bring your own Claude Code or Codex.flockbay.comAd

Glama MCP Gateway

Add one secure layer between your agents and this server.

Resources

Need Help?