Save a report for your owned support_ref in the private AdsAgent support queue. Accepts bounded error classifications, no external recipient. Does not retry or change an advertising operation
| Name | Required | Description | Default |
|---|---|---|---|
| request | Yes | ||
| connection_context | No |
| Name | Required | Description | Default |
|---|---|---|---|
| message | No | ||
| retryable | No | ||
| support_ref | No |
Changes observed during successful MCP inspections.
Input schema / properties / request / additionalPropertiesAdded value: +falseInput schema / properties / request / propertiesAdded value: +{
+ "client": {
+ "additionalProperties": false,
+ "properties": {
+ "guide_version": {
+ "maxLength": 16,
+ "minLength": 12,
+ "pattern": "^\\d{4}-\\d{2}-\\d{2}\\.\\d{1,4}$",
+ "type": "string"
+ },
+ "name": {
+ "enum": [
+ "chatgpt",
+ "claude",
+ "codex",
+ "cursor",
+ "other"
+ ],
+ "type": "string"
+ },
+ "skill_pack_version": {
+ "maxLength": 80,
+ "minLength": 1,
+ "pattern": "^(?:[vV]?\\d+(?:\\.\\d+){0,3}(?:[-+][A-Za-z0-9.-]{1,32})?|git:[0-9a-fA-F]{7,40})$",
+ "type": "string"
+ },
+ "version": {
+ "maxLength": 80,
+ "minLength": 1,
+ "pattern": "^(?:[vV]?\\d+(?:\\.\\d+){0,3}(?:[-+][A-Za-z0-9.-]{1,32})?|git:[0-9a-fA-F]{7,40})$",
+ "type": "string"
+ }
+ },
+ "type": "object"
+ },
+ "idempotency_key": {
+ "maxLength": 128,
+ "minLength": 1,
+ "pattern": "^[A-Za-z0-9][A-Za-z0-9._:/+-]{0,127}$",
+ "type": "string"
+ },
+ "locale": {
+ "maxLength": 35,
+ "minLength": 2,
+ "pattern": "^[A-Za-z]{2,8}(?:-[A-Za-z0-9]{1,8}){0,3}$",
+ "type": "string"
+ },
+ "observation_code": {
+ "enum": [
+ "agent_stopped_fail_closed",
+ "contract_failed",
+ "dependency_failed",
+ "partial_result_unusable",
+ "suspected_regression",
+ "unexpected_server_error",
+ "verification_inconsistent"
+ ],
+ "type": "string"
+ },
+ "submission_mode": {
+ "enum": [
+ "automatic",
+ "manual"
+ ],
+ "type": "string"
+ },
+ "support_ref": {
+ "maxLength": 27,
+ "minLength": 27,
+ "pattern": "^merr_[A-Za-z0-9_-]{22}$",
+ "type": "string"
+ },
+ "task_stage": {
+ "enum": [
+ "setup",
+ "discovery",
+ "read",
+ "prepare",
+ "approval",
+ "confirm",
+ "verification",
+ "reconciliation",
+ "export",
+ "upload"
+ ],
+ "type": "string"
+ },
+ "user_impact": {
+ "enum": [
+ "none",
+ "degraded",
+ "partial_result",
+ "operation_blocked",
+ "write_outcome_uncertain"
+ ],
+ "type": "string"
+ }
+}Input schema / properties / request / requiredAdded value: +[
+ "support_ref",
+ "idempotency_key",
+ "submission_mode",
+ "task_stage",
+ "user_impact"
+]Output schema / else / then / properties / categoryAdded value: +{
+ "type": "string"
+}Output schema / else / then / properties / expectedAdded value: +{
+ "type": "boolean"
+}Output schema / else / then / properties / failure_kindAdded value: +{
+ "type": "string"
+}Output schema / else / then / properties / responsibility_hintAdded value: +{
+ "type": "string"
+}Output schema / else / then / properties / routeAdded value: +{
+ "type": "string"
+}Output schema / else / then / properties / user_visibleAdded value: +{
+ "type": "boolean"
+}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 declare readOnlyHint=false, openWorldHint=false, destructiveHint=false. The description adds genuinely non-obvious behavioral context: it is a write that stays internal ('no external recipient') and has no side effects on ad operations ('does not retry or change'). It does not, however, mention idempotency behavior despite requiring an idempotency_key.
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?
Three tightly packed sentences, purpose front-loaded, no filler. Slightly dense but all content earns its place.
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 no explanation, and annotations cover the safety profile. However, for a tool with a deeply nested required request object at 0% schema description coverage, the description leaves most required parameters unexplained, which is a real gap.
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% and the request object has five required nested fields (support_ref, idempotency_key, submission_mode, task_stage, user_impact) plus optional client/locale/observation_code. The description only clarifies support_ref ownership and gestures at 'bounded error classifications'; it never explains idempotency_key, submission_mode, task_stage, or user_impact semantics, so it fails to compensate for the coverage gap.
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+resource ('Save a report for your owned support_ref') and a precise destination ('private AdsAgent support queue'). This clearly distinguishes it from the sibling support_get_report_status, which only reads status.
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?
Gives clear context for use (bounded error classifications, private/internal, no external recipient) and negative guidance ('Does not retry or change an advertising operation'). It does not explicitly name the alternative for checking a prior report, but the when/when-not framing is otherwise explicit.
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.