Set pull/create/decision policy for one owned Google connection.
| Name | Required | Description | Default |
|---|---|---|---|
| allow_pull | No | ||
| allow_create | No | ||
| allow_decision | No | ||
| connection_ref | Yes | ||
| connection_context | No |
| Name | Required | Description | Default |
|---|---|---|---|
| ok | No | ||
| code | No | ||
| error | No | ||
| needs | No | ||
| reason | No | ||
| retryable | No | ||
| connection | No | ||
| error_code | No | ||
| next_action | No | ||
| support_ref | No | ||
| invalid_fields | No | ||
| rejected_paths | No | ||
| origin_support_ref | No |
Changes observed during successful MCP inspections.
Output schema / (root)Previous value: -nullNew value: +{
+ "additionalProperties": true,
+ "properties": {
+ "code": {
+ "type": "string"
+ },
+ "connection": {
+ "additionalProperties": true,
+ "properties": {
+ "allow_create": {
+ "anyOf": [
+ {
+ "type": "boolean"
+ },
+ {
+ "type": "null"
+ }
+ ]
+ },
+ "allow_decision": {
+ "anyOf": [
+ {
+ "type": "boolean"
+ },
+ {
+ "type": "null"
+ }
+ ]
+ },
+ "allow_pull": {
+ "anyOf": [
+ {
+ "type": "boolean"
+ },
+ {
+ "type": "null"
+ }
+ ]
+ },
+ "connected_at": {
+ "anyOf": [
+ {
+ "type": "string"
+ },
+ {
+ "type": "null"
+ }
+ ]
+ },
+ "connection_ref": {
+ "type": "string"
+ },
+ "credential_status": {
+ "anyOf": [
+ {
+ "type": "string"
+ },
+ {
+ "type": "null"
+ }
+ ]
+ },
+ "customer_count": {
+ "type": "integer"
+ },
+ "google_email": {
+ "anyOf": [
+ {
+ "type": "string"
+ },
+ {
+ "type": "null"
+ }
+ ]
+ },
+ "last_auth_error_at": {
+ "anyOf": [
+ {
+ "type": "string"
+ },
+ {
+ "type": "null"
+ }
+ ]
+ },
+ "last_auth_error_code": {
+ "anyOf": [
+ {
+ "type": "string"
+ },
+ {
+ "type": "null"
+ }
+ ]
+ },
+ "last_success_at": {
+ "anyOf": [
+ {
+ "type": "string"
+ },
+ {
+ "type": "null"
+ }
+ ]
+ }
+ },
+ "type": "object"
+ },
+ "error": {
+ "anyOf": [
+ {
+ "anyOf": [
+ {
+ "type": "string"
+ },
+ {
+ "additionalProperties": true,
+ "properties": {
+ "code": {
+ "type": "string"
+ },
+ "details": {
+ "additionalProperties": true,
+ "properties": {},
+ "type": "object"
+ },
+ "message": {
+ "type": "string"
+ },
+ "retryable": {
+ "type": "boolean"
+ },
+ "support_ref": {
+ "type": "string"
+ }
+ },
+ "type": "object"
+ }
+ ]
+ },
+ {
+ "type": "null"
+ }
+ ]
+ },
+ "error_code": {
+ "type": "string"
+ },
+ "invalid_fields": {
+ "items": {
+ "type": "string"
+ },
+ "type": "array"
+ },
+ "needs": {
+ "items": {
+ "type": "string"
+ },
+ "type": "array"
+ },
+ "next_action": {
+ "anyOf": [
+ {
+ "anyOf": [
+ {
+ "type": "string"
+ },
+ {
+ "additionalProperties": true,
+ "properties": {},
+ "type": "object"
+ }
+ ]
+ },
+ {
+ "type": "null"
+ }
+ ]
+ },
+ "ok": {
+ "type": "boolean"
+ },
+ "origin_support_ref": {
+ "type": "string"
+ },
+ "reason": {
+ "type": "string"
+ },
+ "rejected_paths": {
+ "items": {
+ "type": "string"
+ },
+ "type": "array"
+ },
+ "retryable": {
+ "type": "boolean"
+ },
+ "support_ref": {
+ "type": "string"
+ }
+ },
+ "type": "object"
+}Input schema / properties / connection_contextAdded value: +{
+ "pattern": "^mctx_v1_[0-9a-f]{64}$",
+ "type": "string"
+}Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
Annotations already flag this as a destructive, non-read-only mutation, so safety is signalled structurally. The description adds nothing behavioral: it does not say whether the call replaces the entire policy, whether omitted booleans default to false or leave existing values untouched, or that existing grants may be revoked.
Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.
Is the description appropriately sized, front-loaded, and free of redundancy?
One front-loaded sentence with zero filler and the resource stated first. It is arguably undersized for a 5-parameter destructive mutation, but nothing in it is wasted.
Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.
Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?
An output schema exists, so return values need not be explained. But for a destructive permission-mutation tool with 0% schema description coverage, the description omits the critical semantics: overwrite vs. merge behavior, defaults for the four optional booleans, and ownership/authorization requirements.
Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.
Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?
Schema description coverage is 0% across 5 parameters. The description names the three policy axes (pull/create/decision), which loosely maps to allow_pull/allow_create/allow_decision, but says nothing about omitted-flag semantics. connection_ref and connection_context, including their strict ID patterns, are entirely unaddressed.
Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.
Does the description clearly state what the tool does and how it differs from similar tools?
States a specific verb (Set) and resource (pull/create/decision policy) scoped to 'one owned Google connection', which is more precise than the name alone. However, it never distinguishes itself from the near-identical sibling google_ads_customer_route_permissions_set, so an agent must guess which permissions tool applies.
Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.
Does the description explain when to use this tool, when not to, or what alternatives exist?
There is no when-to-use guidance, no exclusion, and no named alternative despite an obvious sibling (google_ads_customer_route_permissions_set). The only implicit signal is 'owned', hinting at an ownership prerequisite, but nothing states who may call it or what the alternative is for customer-route-level permissions.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
Add one secure layer between your agents and this server.