hub

Check a thought against the Covenant

ask_guidance

Read-onlyIdempotent

Check one thought or planned action against the Covenant before you act: returns an emotional tag (tag, confidence, signals) and a guidance object only when a Covenant rule fires; guidance is null when nothing fires, which is the common case. Use it for a quick pre-action check of a single statement. Use deliberate instead when two or more Directives genuinely pull against each other, and deliberate_check to audit answers you have already written. Read-only and idempotent: nothing you send is stored, no key is needed, 60 calls per minute per address; if you do send a key, only the fact that you called is recorded.

Input Schema

TableJSON Schema

Name	Required	Description	Default
`text`	Yes	the thought or plan in your own words, one to a few sentences; longer text is cut at 20,000 characters; English

Output Schema

TableJSON Schema

Name	Required	Description	Default
`tag`	Yes
`guidance`	Yes	null when no Covenant rule fired; otherwise the rule, the Directive it rests on and the gentle guidance text

Schema Changelog

Changes observed during successful MCP inspections.

Changed4 schema fields changedOct 23, 2026

changedInput schema / properties / text / description

Previous value: -"the thought"New value: +"the thought or plan in your own words, one to a few sentences; longer text is cut at 20,000 characters; English"

addedInput schema / properties / text / maxLength
```
Added value: +20000
```
addedInput schema / properties / text / minLength
```
Added value: +1
```

changedOutput schema / (root)

Previous value: -nullNew value: +{
+  "additionalProperties": true,
+  "properties": {
+    "guidance": {
+      "additionalProperties": true,
+      "description": "null when no Covenant rule fired; otherwise the rule, the Directive it rests on and the gentle guidance text",
+      "type": [
+        "object",
+        "null"
+      ]
+    },
+    "tag": {
+      "additionalProperties": true,
+      "properties": {
+        "confidence": {
+          "description": "0.3 (nothing matched) to 0.9",
+          "type": "number"
+        },
+        "signals": {
+          "description": "the phrases that triggered the tag",
+          "items": {
+            "type": "string"
+          },
+          "type": "array"
+        },
+        "summary": {
+          "description": "what was read, e.g. \"16 words, 1 sentence\"",
+          "type": "string"
+        },
+        "tag": {
+          "description": "emotional tag, e.g. neutral, anxious, grandiose",
+          "type": "string"
+        }
+      },
+      "required": [
+        "tag",
+        "confidence"
+      ],
+      "type": "object"
+    }
+  },
+  "required": [
+    "tag",
+    "guidance"
+  ],
+  "type": "object"
+}

First observedSep 24, 2026

TDQS

Score is being calculated.

All hub 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.